Yale Law School

Yale Law School Legal Scholarship Repository Faculty Scholarship Series

Yale Law School Faculty Scholarship

1-1-1996

Foreword: Sixth Amendment First Principles Akhil Reed Amar Yale Law School

Follow this and additional works at: http://digitalcommons.law.yale.edu/fss_papers Part of the Law Commons Recommended Citation Amar, Akhil Reed, "Foreword: Sixth Amendment First Principles" (1996). Faculty Scholarship Series. Paper 937. http://digitalcommons.law.yale.edu/fss_papers/937

This Article is brought to you for free and open access by the Yale Law School Faculty Scholarship at Yale Law School Legal Scholarship Repository. It has been accepted for inclusion in Faculty Scholarship Series by an authorized administrator of Yale Law School Legal Scholarship Repository. For more information, please contact [email protected].

FOREWORD Sixth Amendment First Principles AKHIL REED AMAR*

In all criminal prosecutions, the accused shall enjoy the right to a speedy and public trial, by an impartial jury of the State and district wherein the crime shall have been committed, which district shall have been previously ascertained by law, and to be informed of the nature and cause of the accusation; to be confronted with the witnesses against him; to have compulsory process for obtaining witnesses in his favor, and to have the Assistance of Counsel for his defence. 1 INTRODUCrION

The Sixth Amendment is the heartland of constitutional criminal procedure, yet the legal community lacks a good map of its basic contours, a good sense of its underlying ecosystem, a good plan for its careful cultivation. Amidst all the Amendment's tightly configured clauses, scholars, lawyers, and judges have often lost their way. The result, at times, has been bad constitutional law and bad criminal procedure. In this article, I offer a general framework for understanding the Sixth Amendment's first principles-for seeing how its many clauses fit together and cohere with other constitutional clauses and principles outside the Amendment. In both interpretive methodology and substantive conclusion, my analysis today dovetails with the analysis put forth in two earlier articles, Fourth Amendment First Principles2 and Fifth Amendment First 3 Principles.

In Part I of what follows, I sketch the major outlines of my project, identifying basic premises and foreshadowing broad conclusions. In Parts II, III, and IV, I get down to details in analyzing the Sixth Amendment's guarantees of speedy trials, public trials, and fair trials, respectively.

* Southmayd Professor, Yale Law School. Special thanks to Bruce Ackerman, Peter Arenella, Jon Blue, Paul Cassell, Richard Friedman, Al Hirsch, Neal Katyal, Nancy King,

John Langbein, Ren~e Lettow, Amit Saluja, Steve Saltzburg, David Sklansky, Bill Stuntz, Jon Varat, and Ron Wright. 1. U.S. CONST. amend. VI. 2. See Akhil Reed Amar, Fourth Amendment First Principles, 107 HIRv. L. REv. 757

(1994) [hereinafter Amar, Fourth Amendment]. 3. See Akhil Reed Amar & Ren~e B. Lettow, Fifth Amendment First Principles: The Self-Incrimination Clause, 93 MICH. L. REv. 857 (1995) [hereinafter Amar & Lettow, Fifth Amendment].

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

I. IN GENERAL: FRAMING THE ISSUES

To illuminate the internal architecture of the Sixth Amendment, I shall organize my account around three clusters of rights. First is the basic right to a speedy trial, a right embodied in a single clause that, as we shall see, in fact protects a cluster of distinct interests, including (a) a physical liberty interest in avoiding prolonged pretrial detention; (b) a mental liberty and reputational interest in minimizing unjust accusation; and (c) a reliability interest in assuring that the accuracy of the trial itself is not undermined by an extended accusation period. After this speedy trial cluster comes a cluster of rights to a public trial-a trial of, by, and before the people. In a republican government, a trial should be a res publica, a public thing, the people's thing. Included in this cluster are the rights to (a) a trial held in public, (b) featuring an impartial jury of the people, (c) who come from the community where the crime occurred. Finally comes the cluster offair trial rights, encompassing notice and the opportunity to hear and be heard. Put slightly differently, this last cluster safeguards the right to know, and defend oneself against, an accusation of criminal wrongdoing. Textually, this cluster encompasses the rights to (a) be informed of the nature and cause of accusation; (b) be confronted with prosecution witnesses; (c) compel the production of defense witnesses; and (d) enjoy the assistance of counsel in defending against the accusation. The deep principles underlying the Sixth Amendment's three clusters and many clauses (and, I submit, underlying constitutional criminal procedure generally) are the protection of innocence and the pursuit of truth. The speedy trial right protects the innocent man from prolonged de facto punishment--extended accusations that limit his liberty and besmirch his good name-before he has had a fair chance to defend himself. If government accuses an innocent man and then refuses to suspend its accusation, it must give him the right, speedily, to clear himself at trial and regain his good name and full liberty. And if government holds the accused in extended pretrial detention, courts must ensure that the accuracy of the trial itself will not thereby be undermined-as might occur if an innocent defendant's prolonged detention itself causes the loss of key exculpatory evidence. So too, the public trial right protects the innocent man from an erroneous verdict of guilt. Witnesses for the prosecution may be less willing to lie or shade the truth with the public looking on; and bystanders with knowledge of the underlying events can bring missing information to the attention of court and counsel. A defendant will be convicted only if the people of the community (via the jury) believe the criminal accusation-believe both that he did the acts he is accused of, and that these acts are indeed criminal and worthy of the community's moral condemnation. This last

1996]

SIXTH AMENDMENT FIRST PRINCIPLES

aspect-passing judgment on a defendant's normative guilt or innocence-is an especially important part of the public trial idea. Finally, the fair trial right also protects the innocent man from an erroneous verdict of guilt, though its safeguards highlight factual innocence ("I didn't do it") more than normative innocence ("I did it, but I did not thereby offend the public's moral code"). Counsel, confrontation, and compulsory process are designed as great engines by which an innocent man can make the truth of his innocence visible to the jury and the public. To say, as I do, that the Sixth Amendment is generally designed to elicit truth and protect innocence' might at first seem either dangerous or trivial. If my reading of the Amendment protects only innocent men and women, it would indeed be dangerous-surely the Amendment protects all accused persons, the guilty along with the innocent, in affirming rights to speedy, public, and fair trials. If, alternatively, my reading of the Amendment concedes this obvious point, it might seem to border on the trivial: if the Amendment protects both the guilty and innocent, how can it be said, in any deep or interesting way, to be about innocence? Who could be against the (trivial) idea that innocent people have rights too-the same rights as the guilty? The above dilemma, I submit, is a false one. My account of the Amendment is neither totalitarian nor trivial. Many parts of the Amendment, rightly read, do not protect only innocents, but they do protect only innocence; they protect the guilty only as an incidental by-product of protecting the innocent because of their innocence.' Put another way, although the guilty will often have the same rights as the innocent,6 they should never have more, and never because they are guilty.

4. I say "generally" because, as we shall see, the Sixth Amendment also protects other values, such as popular sovereignty and republican political participation-values that in general complement rather than contradict innocence protection and truth-seeking. 5. Consider, for example, the Winship due process principle, which requires proof beyond reasonable doubt in criminal cases. See In re Winship, 397 U.S. 358, 362-64 (1970). Though the rule will have the incidental statistical effect of freeing some guilty defendants, the purpose of the rule is obviously to protect the innocent defendant from erroneous conviction. Although many accused persons are indeed guilty, we cannot know which ones before reliable Sixth Amendment trials have occurred. A person who is, at the time of the crime, factually and normatively guilty is legally presumed innocent until proved and found guilty; and until then, the guilty defendant incidentally benefits from Sixth Amendment rules designed to protect innocent defendants from erroneous convictions. 6. At times, guilty defendants should enjoy less freedom than do innocent ones. For example, a guilty defendant should at times be less free to try to demolish, via crossexamination, a truthful witness than would an innocent defendant facing a lying witness. See infra Part IVD. So too, when it comes to remedies, the guilty may at times recover less than the innocent because, as we shall see, the guilty may have suffered less constitutionally

cognizable legal injury. For example, if guilty A endures one month of unlawful pretrial detention, but upon conviction gets a one month sentencing discount for time served, A has suffered less cognizable injury than innocent B who endures the same unlawful pretrial detention and is then acquitted. See generally infra Part II.

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

These last points, too, might seem trivial to ordinary Americans-they reflect common sense-but they sharply conflict with various doctrines of modern constitutional criminal procedure that many judges and welltrained lawyers take for granted these days. These modern doctrines create what I shall call an upside-down effect, providing the guilty with more protection than, and often at the expense of, the innocent. For example, our Fourth Amendment caselaw at times has suggested that criminal suspects receive more privacy protection than presumptively lawabiding citizens:7 exceptions to the so-called probable cause and warrant requirements are apparently easier to justify when the government is not seeking evidence from criminal suspects but is instead intruding on 8 privacy interests of individual members of the general public. Yet nothing in the text, history, or structure of the Fourth Amendment supports such 9 an upside-down approach to privacy rights. On the remedy side of the Fourth Amendment, caselaw is likewise upside down. The exclusionary rule creates huge windfalls for guilty defendants,'" but gives no direct remedy to the innocent woman wrongly searched. The guiltier you are, the more evidence the cops find, the bigger the exclusionary rule windfall; but if the cops know you are innocent and just want to hassle you (because of your race, or politics, or whatever), the exclusionary rule offers exactly zero deterrence or compensation. Here too, nothing in the Fourth Amendment's text, history, or structure supports such an upside-down and truthsuppressing remedial scheme." Current interpretations of the Fifth Amendment's Self-Incrimination Clause are likewise upside down. Courts and commentators dwell on the so-called "cruel trilemma" of self-accusation, perjury, or contempt faced by some defendants.' 2 But this classic trilemma arises only if a person is in fact guilty. (Otherwise he need not directly accuse himself by speaking truthfully, and commits no perjury when he asserts his innocence.) Why is the trilemma so "cruel" if one can avoid it simply by not committing crimes? By contrast, courts and commentators have often overlooked the distinctively cruel choice faced by some innocent defendants who, if forced 7. At other times, however, the Supreme Court has rejected and even inverted this premise. See infra text accompanying notes 110-13. 8. See Amar, FourthAmendment, supra note 2, at 801. 9. See id. at 758, 770, 801. 10. It is often argued that no exclusionary rule windfall exists because, if the government had never violated the Fourth Amendment, it never would have gotten the evidence in the first place. Thus (the argument goes), the exclusionary rule creates no windfall, but simply restores the status quo ante. This argument is slick, but wrong. It ignores what I have elsewhere called the "causation gap"-in many situations the government would have ultimately found the evidence or some substitute even if no constitutional violation ever occurred. For elaboration and analysis, see generally Amar, FourthAmendment, supra note 2, at 793-95; Amar & Lettow, Fifth Amendment, supra note 3. 11. See Amar, FourthAmendment, supra note 2, at 785-800.

12. See Amar & Lettow, Fifth Amendment, supra note 3, at 890.

19961

SIXTH AMENDMENT FIRST PRINCIPLES

to take the stand, might (say, because of nerves or an offputting manner) hurt their own cause and be erroneously convicted. This upside-down account of the cruel trilemma has had a huge effect in self-incrimination law. Its misplaced tenderness towards the guilty has led courts to needlessly exclude, in the name of Self-Incrimination Clause values, reliable physical evidence of guilt 3 -evidence that is in no sense Fifth Amendment "witness[ing]." This exclusion is a windfall to the guilty without any offsetting benefit for the innocent. Even worse, an overbroad reading of Fifth Amendment self-incrimination has led courts to deny an innocent defendant her explicit Sixth Amendment right to compulsory process against a guilty witness who asserts his own right to avoid the cruel trilemma. Our innocent defendant knows who committed the crime, but today she cannot force him to take the stand in her own trial, even though her liberty and good name-perhaps even her life-are on the line. 4 Here too, nothing in the Constitution, rightly read, supports this upside-down and truth-suppressing effect.5 In two earlier articles, I documented and critiqued these upside-down effects in current interpretations of the Fourth and Fifth Amendments, respectively; and in this article I propose to do the same thing for the Sixth Amendment. The Sixth Amendment Speedy Trial Clause is my Exhibit A. Perhaps influenced by misguided Fourth and Fifth Amendment doctrines excluding reliable evidence of guilt, the Supreme Court, in the name of the Speedy Trial Clause, has created the mother of all upside-down exclusionary rules. "The only possible remedy" for speedy trial violations, the Court has unanimously proclaimed, is dismissal of the case with prejudice-in effect, excluding all evidence of guilt forever.1 6 At first blush, the Court's pronouncement seems plausible: if too much time has already elapsed, how can the government ever hold a constitutionally proper trial in the future? But as we shall see, this initial reaction is wrong in just about every way imaginable. As a matter of logic, there are many other "possible" remedies. As a matter of general remedial theory, dismissal with prejudice is rarely the remedy that best "fits" the legal rights and interests that have been violated. As a matter of history, some alternative remedies have deep roots in the common law underlying speedy trial. (The concept of dismissal with prejudice as the "only possible remedy" has no such roots.) As a matter of text and structure, dismissal with prejudice makes no sense as a response to many types of speedy trial violations. And as a matter of precedent, the modern Supreme Court has said and done many things that

13. See, e.g., Kastigar v. United States, 406 U.S. 441 (1972). 14. See Amar & Lettow, Fifth Amendment, supra note 3, at 861-64. 15. See generally id., passim. 16. See Strunk v. United States, 412 U.S. 434, 440 (1973); Barker v. Wingo, 407 U.S. 514, 522 (1972).

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

are logically inconsistent with what it has said and done about dismissal with prejudice. The speedy trial dismissal remedy provides a windfall for the guilty while leaving the innocent defendant, who has suffered excessive detention or unjustified stigma owing to an extended accusation, uncompensated. (In this respect, dismissal resembles the Fourth Amendment exclusionary rule.) But dismissal is even more upside down than this in practice. Because judges (rightly) see the remedy as extreme, they are loath in any given case to admit that the speedy trial right was indeed violated. As a result, many innocent defendants are made affirmatively worse off, suffering greater violations of their explicit constitutional rights. (In this respect, dismissal resembles current doctrine under the Fifth Amendment SelfIncrimination Clause.) A sensible constitutional criminal procedure, I submit, must systematically right upside-down effects in current Fourth, Fifth, and Sixth Amendment doctrine. It must also begin to take constitutional text seriously. In previous work I have argued that the words of the Fourth Amendment really do mean what they say.17 They do not require warrants or probable cause for all searches and seizures, but they do require that all searches and seizures be reasonable. The words do not require exclusion of reliable evidence in criminal trials, but they do presuppose common law and other property and tort law remedies that secure Americans in their "persons, houses, papers, and effects." So too, I have argued that the words of the Fifth Amendment mean what they say. The words "same offence" in the Double Jeopardy Clause really do mean "same offence," rather than "greater and lesser-included offences" or "same factual transaction" or any number of other things.18 The Self-Incrimination Clause really means that a criminal defendant must not be forced to be a "witness" against himself "in" a "criminal case"-by taking the stand at trial or having a compelled out-of-court affidavit or transcript introduced. 1 9 But the clause does not say that if a person is forced to be a witness against himself before Congress, or in a civil case, or anywhere else outside his own 17. See U.S. CONST. amend. IV ("The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized."). See generally Amar, FourthAmendment, supra note 2. 18. See U.S. CONST. amend. V ("[N]or shall any person be subject for the same offence to be twice put in jeopardy of life or limb .... ). See generally Akhil Reed Amar & Jonathan Marcus, Double Jeopardy Law After Rodney King, 95 COLUM. L. REV. 1 (1995). A plain meaning reading of the Double Jeopardy Clause, however, must be supplemented by basic due process principles prohibiting vexatious or innocence-threatening multiple prosecutions generally. See id. at 28-38. 19. See U.S. CONST. amend. V ("No person ... shall be compelled in any criminal case to be a witness against himself .... ). See generally Amar & Lettow, Fifth Amendment, supra note 3.

1996]

SIXTH AMENDMENT FIRST PRINCIPLES

criminal case, the fruits of that witnessing must be excluded from his criminal case. Unless the witnessing itself occurs inside his criminal case-in person or by affidavit or transcript-the words of the Self-Incrimination Clause, and its innocence-protecting spirit, are satisfied. A similar attention to the word "witness" will neatly solve many of the problems that currently beset Sixth Amendment Confrontation Clause doctrine. The modern Court has viewed the clause as implicated whenever hearsay comes into the criminal courtroom: If in-court A testifies about 20 what out-of-court B said, the defendant's right to confront B is at stake. But surely all hearsay cannot be unconstitutional. At common law, the traditional hearsay "rule" was notoriously un-ruly, recognizing countless exceptions to its basic preference for live testimony; and more recent statutes have proliferated exceptions. But the words and grammar of the Confrontation Clause are emphatically rule-ish: "In all criminal prosecutions, the accused shall enjoy the right ...

to be confronted with the

witnesses against him"-no ifs, ands, or buts. And so the modern Court has put itself in a bind. If the clause does truly prohibit all hearsay, as its grammar might imply, it is utterly unworkable; but to make it workable-by recognizing commonsensical exceptions-is to offend its seeming grammar. The obvious solution is to heed the word "witness" and its ordinary, everyday meaning. If I tell my mom what I saw yesterday, and she later testifies in court, I am not the witness; she is. Not all out-of-court declarants within the meaning of the so-called hearsay rule are "witnesses" within the meaning of the Confrontation Clause. In the Fifth Amendment SelfIncrimination Clause, "witness" means a person who physically takes the stand to testify, or (to prevent government evasion of the spirit of the clause) a person whose out-of-court affidavit or deposition (prepared by the government for in-court use) is introduced as in-court testimony. In the Sixth Amendment the word "witness" means the same thing, and for the same reason. Once we see this, the Court's current Confrontation Clause conundrum vanishes. The clause means what it says, and the strict rule it lays down makes sense as a rule. A sensible Sixth Amendment jurisprudence must begin with plain meaning, but it must not end there. Though the rules of the Amendment make sense as rules, deeper principles lurk beneath the rules. The Amendment does mean what it says; but sometimes it means even more. In many contexts, the expressio unius maxim is a sound one, but in the Sixth Amendment we must not apply the maxim woodenly. 21 The Amendment

20. See, e.g., Maryland v. Craig, 497 U.S. 836, 849 (1990); Ohio v. Roberts, 448 U.S. 56, 63 (1980). 21. The maxim expressio unius est exclusio alterius means that the expression of one thing by implication excludes other things.

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

recognizes that the accused has a right to a public trial, but perhaps the public trial right is also a right of, well, the public itself-the people. Likewise, the Amendment vests the accused with a right to a jury, but surely the people themselves have a jury right too. Article III says that "[t]he Trial of all Crimes... shall be by Jury, ' 22 whether an accused who pleads not guilty wants one or not;2 3 and nothing in the words or history of the Sixth Amendment reveals any purpose to repeal that clear command. 4 More generally, of course, the Ninth Amendment explicitly tells us not to infer by expressio unius that a "right[] ...[of] the people" has been surrendered. 25 And this explicit reminder seems especially apt when we deal with what are quite literally rights of "the people"-rights, that is, of the public and populace at large.2 6 But the Ninth Amendment reminder must radiate more broadly than this when we read the Sixth Amendment, lest we reach absurd results. The Confrontation Clause says that the accused has a right to observe and examine the government's witnesses, but surely the accused must also have a right to observe and examine the government's physical evidence, although the Amendment does not explicitly say so. The Compulsory Process Clause affirms the defendant's right to forcibly subpoena a witness; but surely the Constitution must also protect the defendant's right to present friendly witnesses who volunteer to testify on his behalf, although the Amendment again does not explicitly say so. 27 If the defendant has a Sixth Amendment right to present exculpatory witnesses to the jury, surely he must also have a right to present exculpatory physical evidence, although, here too, the Amendment does not explicitly say so. If we insist on being textualists and only textualists, we can hide behind 28 the explicit texts of the Ninth Amendment and the Due Process Clause

3. 22. U.S. CONST. art. III, § 2, cl. 23. If the accused pleads guilty, there is, strictly speaking, nothing to try, and no trial. See Akhil Reed Amar, The Bill of Rights as a Constitution, 100 YALE LJ.1131, 1199 & n.301 (1991).

24. Contra Patton v. United States, 281 U.S. 276 (1930). For discussion and criticism of Patton, see Amar, supra note 23, at 1196-99.

25. See U.S. CONST. amend. IX ("The enumeration in the Constitution, of certain rights, shall not be construed to deny or disparage others retained by the people."). 26. For a similar invocation of the Ninth Amendment to affirm the people's right to a public trial, and to rebut an expressio unius reading of the Sixth Amendment reference to a right of "the accused," see Richmond Newspapers, Inc. v. Virginia, 448 U.S. 555, 579 n.15 (1980) (plurality opinion). 27. A clever textualist might note implicit textual support from the juxtaposition of the Confrontation Clause, which speaks of "the witnesses" for the prosecution, and the Compulsory Process Clause, which speaks only of the defendant's right to subpoena "witnesses"not "the witnesses"-he plans to present. This juxtaposition implicitly points to the existence of other witnesses-presumably nonsubpoenaed-that the defendant might want to put on. But even this juxtaposition does not explicitly make clear that the defendant has a constitutional, as opposed to a possible statutory or common law, right to put on witnesses who volunteer. 28. See U.S. CONST. amend. V ("No person shall be ... deprived of life, liberty, or

1996]

SIXTH AMENDMENT FIRST PRINCIPLES

to fill in the obvious textual gaps in the Sixth Amendment. But this hypertextual strategy misses how the rules of the Sixth Amendment themselves should influence sensible Due Process analysis. Behind the words of the Sixth Amendment rules are indeed "postulates which limit and control, '29 "emanations" and "penumbras, ' - 30 spirit and structure as well as text-in short, Sixth Amendment principles as well as Sixth Amendment rules. And the first principles underlying the rules are, I submit, the protection of innocence and the commitment to truth-seeking trials. These first principles, of course, explain why it seems so obvious that a defendant must have a right to defend himself in certain ways not explicitly covered by the words of the Confrontation and Compulsory Process Clauses. Protecting the innocent, pursuing the truth, and respecting the textthese, I claim, are the basic elements of a sensible Sixth Amendment jurisprudence (and, more generally, a sensible jurisprudence of constitutional criminal procedure). To see more clearly how this jurisprudence might work, and what it would entail in the Sixth Amendment, we need to get specific. II. IN PARTICULAR: SPEEDY TRIAL The Supreme Court has said and done a great deal about the Speedy Trial Clause in the last three decades: (1) It has repeatedly identified three major and distinct interests protected by the clause-an interest in avoiding prolonged pretrial detention, an interest in minimizing the anxiety and loss of reputation accompanying public accusation, and an interest in assuring the ultimate fairness of a long-delayed trial.3 1 (2) It has made clear that the "major evils" of pretrial restraints on liberty and loss of reputation occasioned by accusation "exist quite apart from actual or possible prejudice to an accused's defense. 3 2 (3) It has held that the clause, by its plain meaning, simply does not apply to the time period between the commission of a crime and the date when a person is "accused" by the government (typically via arrest or indictment).3 3 (4) Likewise, it has held that when a person ceases to be "accused"-because the government formally drops charges while retaining the right to reindict later-this nonaccusation period does not count against the government

property, without due process of law .... "); see also id. amend. XIV, § 1 ("[Nlor shall any State deprive any person of life, liberty, or property, without due process of law ... . 29. Monaco v. Mississippi, 292 U.S. 313, 322 (1934). 30. Griswold v. Connecticut, 381 U.S. 479, 484 (1965). 31. See United States v. Ewell, 383 U.S. 116, 120 (1966); Smith v. Hooey, 393 U.S. 374, 378 (1969); United States v. Marion, 404 U.S. 307, 320 (1971); Barker v. Wingo, 407 U.S. 514, 532 (1972); United States v. Loud Hawk, 474 U.S. 302, 312 (1986); Doggett v. United States, 505 U.S. 647, 654 (1992). 32. Marion, 404 U.S. at 320.

33. Id. at 313-21.

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

for Speedy Trial Clause purposes.3 4 (5) It has held that if delay during the preaccusation period (and presumably the nonaccusation period during which formal charges are dropped altogether) compromises the defendant's ability to defend herself fairly at trial, the "primary guarantee" against injustice comes from "the applicable statute of limitations," but due process principles may also prevent the trying of egregiously stale charges. 35 (6) It has noted that the judicial remedy of dismissing a case with prejudice for Speedy Trial Clause violations is "unsatisfactorily severe ...because it means that a defendant who may be guilty of a serious crime will go free, without having been tried. Such a remedy is more serious than [the Fourth Amendment] exclusionary rule.",3 6 (7) Nevertheless, it has also said, repeatedly and unanimously, that dismissal with prejudice-that is, dismissal with no possibility of refiling charges later-is "the only possible remedy" for Speedy Trial Clause violations.37 What's wrong with this picture? What's wrong, I submit, is that proposition (7) simply does not follow from, and is in fact logically and practically inconsistent with, propositions (1)-(6). The first six propositions are sound, with firm roots in constitutional text, history, and structure, and in common sense. But proposition (7) betrays all this.3 8 A. PRECEDENTIAL LOGIC AND REMEDIAL THEORY

Consider first some basic premises of general remedial theory as applied to propositions (1)-(6). If, as proposition (1) holds, there are indeed three distinct legal interests underlying the Speedy Trial Clause, it would be odd that, remedially, one size fits all-that only one possible remedy (and an admittedly drastic one at that) exists. Each legal interest has a unique size and shape, and its own uniquely apt remedy package. Remedies should fit rights, and if rights (or "legal interests") do not come in a one-size-fits-all package, neither should remedies. If Andy is arrested on the day of the crime and held in jail pretrial, let's stipulate that Andy's speedy trial right to be free from prolonged pretrial detention will be violated unless he is brought to trial within, say, a month of his incarceration. 3 9 But if Bill is accused of the same crime on the same

34. See United States v. MacDonald, 456 U.S. 1, 6-9 (1982); Loud Hawk, 474 U.S. at 310-12. 35. See Ewell, 383 U.S. at 122; Marion, 404 U.S. at 322-25; United States v. Lovasco, 431 U.S. 783, 789 (1977). 36. Barker, 407 U.S. at 522. 37. Strunk v. United States, 412 U.S. 434, 439-40 (1973); Barker, 407 U.S. at 522. 38. Though I disagree with his analysis at key points in what follows, I have greatly profited from Anthony G. Amsterdam, Speedy Criminal Trial: Rights and Remedies, 27 STAN. L. REV. 525 (1975).

39. This assumes that Andy does not, for example, knowingly and intelligently waive his speedy trial rights by asking for a postponement to better prepare his defense. The precise

19961

SIxTH AMENDMENT FIRST PRINCIPLES

day and is free on his own recognizance pending trial, a delay of more than a month between accusation and trial would not necessarily be unconstitutional. A wholly distinct interest-an interest in minimizing Bill's reputation loss and anxiety caused by public accusation-is at stake, and that interest will not be violated unless the accusation period stretches out for, say, one year. The fair trial legal interest is different still. If Cindy is never even accused of the crime until eighteen months after the event, the Speedy Trial Clause would not generally demand a trial immediately after indictment even though the evidence in Cindy's trial may well be more stale than in Andy's and Bill's trials. This point about speedy trial rights has dramatic implications for speedy trial remedies. If our incarcerated defendant Andy demands a speedy trial on day 30, and the prosecutor is unable to proceed to trial forthwith, a judge could simply order Andy released on his own recognizance, giving the prosecutor (in our hypothetical) eleven more months. Thus, the government will never have violated the Speedy Trial Clause in this situation; Andy's case will become just like Bill's.4 But suppose instead that our trial judge fails to do the right thing on day 30, and she wrongly keeps Andy in detention for, say, five more months and only then releases him from jail. Suppose further that a trial is held four months after release-ten months after the crime, arrest, and initial detentionand that the extra five months of unjustified detention have not irreversibly compromised Andy's ability to defend himself fairly at trial.4 1 Has the Speedy Trial Clause been violated in this scenario? Of course, by hypothesis. Andy's legal interest in avoiding undue pretrial detention was violated by five extra months in jail. But the trial itself did not violate the Speedy Trial Clause, nor did the date of the trial.4 2 The constitutional wrong here is detention, not the trial or its timing; if the judge had released Andy on day cardinalfigure here-one month-is arbitrary, and used merely for illustrative simplicity. As we shall see, my point focuses on the ordinal relation between this number and later numbers in our hypothetical. 40. The speedy trial legal interest in avoiding unduly long pretrial detention will not have been violated because Andy will have spent only the allowable one month in jail. And the speedy trial legal interest in avoiding undue anxiety and loss of reputation because of an extended accusation will not have been violated so long as the "accusation period" is less

than one year. This is the lesson of Bill's case. (This assumes, for illustrative simplicity, that pretrial jail time does not exacerbate the reputation damage caused by criminal accusation simpliciter. If we suspend this assumption, the precise math changes, but not my basic point. Assume for example that one month in jail is, in terms of damage to reputation,equivalent to two months of mere (undetained) accusation. If so, then when Andy is released from jail on day 30, our prosecutor would have only 10 rather than 11 more months of accusation left.) 41. This last assumption is made here only for illustrative simplicity. Later I shall explore this assumption and analyze situations where it does not hold. See infra text accompanying notes 62-65. 42. See United States v. MacDonald, 435 U.S. 850, 861 (1978) ("It is the delay before trial, not the trial itself, that offends against the constitutional guarantee of a speedy trial .... Proceeding with the trial does not cause or compound the deprivation already suffered.").

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

30, the trial could have been held ten months after the crime, arrest, and initial detention without any violation of the Speedy Trial Clause. Andy's case would have been no different from Bill's. Without this timely release, Andy's overlong detention is a constitutional violation regardless of when, or even whether, a trial later takes place. (This is a key implication of proposition (2) as laid down in the Marion case.) 4 3 Undue detention is an unjustified trespass-an unreasonable seizure of the person. This trespass, this unreasonable seizure, is a special outrage if Andy is in fact innocent-if the government would have dropped the charges against him anyway, rather than going to trial, or if the government would have lost the case against him at trial." Dismissal with prejudice is a huge windfall to Andy if guilty-it makes him unjustifiably better off than a guilty Bill-but fails to fully remedy the constitutional wrong perpetrated against Andy if innocent. As a remedy, dismissal with prejudice is thus an inapt, misfitting remedy for the legal interest violated. It is perverse and upside down. (Indeed, it simply magnifies the perversity of the Fourth Amendment exclusionary rule's approach to unreasonable searches and seizures generally, giving the guilty a windfall and the innocent a brushoff.) A constitutionally apt enforcement and remedial regime for Andy's liberty rights will call for injunctive and habeas suits, and framework statutes, to prevent or limit ongoing violations of his bodily liberty; after-the-fact compensatory and (in egregious cases) punitive damages for any unconstitutional detention time actually served; and sentencing offsets (if Andy is ever convicted) for time served to avoid double punishment.4 5 Next, consider the case of Bill-accused the day of the crime but never detained pretrial. If, one year later, Bill moves for an immediate trial and the trial judge grants his motion, no speedy trial right will ever have been violated. But suppose, instead, that our trial judge wrongly denies Bill's motion and he does not get his trial until five months later, seventeen months after initial accusation. Suppose further that the extra five months of undue anxiety and reputation loss that Bill suffers because of the overlong accusation in no way compromise his ability to defend himself fairly at trial.4 6 Has the Speedy Trial Clause been violated in this scenario? Of course, by hypothesis. Bill's legal interest in his good name and peace of mind was violated by five extra months of accusation. But the trial itself 43. See supra text accompanying note 32; United States v. Marion, 404 U.S. 307, 320 (1971) ("[Tlhe major evils protected against by the speedy trial guarantee exist quite apart from actual or possible prejudice to an accused's defense."). 44. See Barker v. Wingo, 407 U.S. 514, 533 (1972) ("Imposing those consequences [of pretrial detention] on anyone who has not yet been convicted is serious. It is especially unfortunate to impose them on those persons who are ultimately found to be innocent."). 45. See infra Part IIB2a. 46. Again, I introduce this assumption here only for ease of exposition. Later, we shall examine this assumption more carefully. See infra text accompanying notes 66-67.

1996]

SIXTH AMENDMENT FIRST PRINCIPLES

653

did not violate the Speedy Trial Clause, nor did the date of the trial.4 7 The constitutional wrong here is overlong accusation and stigma, not the trial or its timing. If the judge had dismissed Bill's accusation after one year, with explicit leave to the prosecutor to reindict whenever she was ready to proceed to trial forthwith, the Sixth Amendment accusation period would have been tolled, and the government-created stigma created by pending accusation would have lasted no more than one year. (This is the clear meaning of proposition (4) as laid down in the MacDonaldand Loud Hawk cases.)4 8 In this scenario, upon reindictment five months later, a trial could have been held seventeen months after initial accusation, without any violation of the Speedy Trial Clause. Bill's case would have been no different from Cindy's.4 9 Without this tolling of accusation via dismissal 47. See supra note 42. 48. See supra text accompanying note 34; United States v. MacDonald, 456 U.S. 1, 6-9 (1982); United States v. Loud Hawk, 474 U.S. 302, 310-12 (1986); see also United States v. Marion, 404 U.S. 307, 320 (1971). As the MacDonald Court noted, Congress, in enacting the Speedy Trial Act of 1974, likewise believed that dropping an indictment would toll the Amendment. See MacDonald, 456 U.S. at 7 n.7. According to MacDonald, a person whose indictment has been dropped, with leave to refile, is in no different position than one who has never been indicted, but who has come under suspicion and is the subject of an ongoing (and perhaps highly publicized) "investigation." See id. at 8-9. Since the speedy trial clock doesn't run preaccusation, neither should it run in nonaccusation intervals, according to MacDonald. See id. at 7. (Even if this view were rejected, the speedy trial clock could be deemed to run-but more slowly-during nonaccusation intervals: preaccusation months don't count, accusation months count fully, and nonaccusation-interval months could count somewhere in between on the theory that the first indictment creates special stigma not wholly dissipated by dismissal.) 49. Suppose instead that Bill's indictment had stretched out 17 years rather than 17 months. In this case we should still compare Bill's suit to a comparably situated Cindy who was indicted for the first time 17 years after the crime. If the case against Cindy would be barred by the applicable statute of limitations, so might the case against Bill. Pending indictments typically toll applicable statutes of limitations. See, e.g., Klopfer v. North Carolina, 386 U.S. 213, 214 (1967); cf Dickey v. Florida, 398 U.S. 30, 32 n.3 (1969) (arrest warrant). But this tolling rule should arguably not apply to any period of unconstitutional indictment-any period after one year in Bill's case. Indeed, after one year, Bill should arguably have a right to treat the indictment as constitutionally lapsed, and to insist on a new indictment. This would insure a rough contemporaneousness of judgment of the grand and petit juries-the two panels of the people who must both pass judgment against a federal defendant before any conviction can occur. This analysis may help explain both the result and some of the language of Justice Souter's opinion for the Court in Doggett v. United States, 505 U.S. 647 (1992), a drug case where more than eight years lapsed between indictment and arrest. Despite a powerful dissent from Justice Thomas distinguishing statute of limitations concerns from Speedy Trial Clause concerns, id. at 667-71 (Thomas, J., dissenting), the Court's interpretation of the clause was influenced by concerns about evidentiary staleness. Id. at 654-56 (opinion of the Court). The relevant statute of limitations for the underlying offense was five years. See United States v. Doggett, 906 F.2d 573, 583 (11th Cir. 1990) (Clark, J., dissenting) (same case). Justice Souter's opinion draws some support from Dickey, 398 U.S. 30, but Dickey was decided prior to Marion, MacDonald,and Loud Hawk; and some of its language and logic do not survive those cases. See, e.g., Dickey, 398 U.S. at 37-38 (apparently linking Sixth Amendment to delay between crime and trial); id. at 40 (Brennan, J.,concurring) (noting that Court leaves open issue of prearrest delay).

654

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

and reindictment, Bill's loss of his good name for five unjustified months is a constitutional violation regardless of when, or even whether, a trial later takes place. (Again, this follows from proposition (2) as laid down in Marion.)5 ° Besmirching an innocent man's good name, while denying him a quick chance to clear himself, is a kind of verbal assault-a reputational mugging.5 Of course, if Bill is in fact guilty as charged, he may lack good ground for complaint. In effect, truth of the accusation may be a defense that renders Bill's injury moot.5 2 (And so, the constitutional violation of guilty Bill's reputation interest during the extra five months might be a species of harmless error.) As with Andy's case, a dismissal with prejudice here would be perverse and upside down. The guilty man gets a windfall and the innocent one gets nothing for five unconstitutional months of mud on his name. A constitutionally apt enforcement and remedial regime for Bill's reputation right will call for timely judicial orders to either prosecute now or drop (for now) the pending accusation, with leave to reindict later--orders that would prevent or limit ongoing assaults on reputation; a requirement (at least for federal defendants) that any delayed trial be preceded by a fresh grand jury reindictment;5 3 and after-the-fact compensatory and (in egregious cases) punitive remedies for innocent defendants who have suffered false and prolonged assaults on their good names.5 4 Consider finally the case of Cindy-accused eighteen months after the crime, and free on her own recognizance until her trial, say, three months later. Suppose at trial Cindy seeks to dismiss her case with prejudice. She claims that her right to defend fairly has been irreversibly harmed by the long time period between crime and trial. Critical exculpatory evidence that was once available has now been lost in the mist of time; key defense witnesses have died or moved, or their memories have faded. But if these things happened in the first eighteen months, preindictment, Cindy's speedy trial claim is a clear loser. Under propositions (3) and (4) as laid down in Marion, MacDonald,and Loud Hawk, the clause applies only to "accused" persons, and simply has no relevance to a preaccusation (and presumably

50. See supra text accompanying note 32; Marion, 404 U.S. at 320.

51. See Klopfer, 386 U.S. at 216 (holding that unconstitutional delay denies accused "an opportunity to exonerate himself'). 52. Cf. 3 WILLIAM BLACKSTONE, COMMENTARIES *125 ("[I]f the defendant be able to justify, and prove the words to be true, no action will lie, even though special damage hath ensued: for then it is no slander or false tale."). 53. See supra note 49. Unlike most of the Bill of Rights, the Fifth Amendment requirement that serious criminal prosecutions be preceded by a grand jury indictment has not been "incorporated" against states via the Fourteenth Amendment. See Beck v. Washington, 369 U.S. 541, 545 (1962); Hurtado v. California, 110 U.S. 516, 538 (1884). For criticism of this, see Akhil Reed Amar, The Bill of Rights and the Fourteenth Amendment, 101 YALE L.J. 1193, 1218-43, 1263-64 (1992).

54. See infra text accompanying notes 109-13.

1996]

SIXTH AMENDMENT FIRST PRINCIPLES

any tolled nonaccusation) period. Even if all the evidence loss occurred in the final three (accusation period) months, Cindy should not yet be home free. If the clause applies only to evils in the accusation period, doesn't logic suggest that it should apply only to evils caused by accusation itself? 56 In other words, doesn't it apply only to harms that occur when and because one is "accused"? If so, Cindy should need to show that the accusation itself helped cause the loss of evidence. If not, Cindy's speedy trial claim should be no different than the claim of a comparably situated defendant-call her Denise-indicted twenty-one months after the crime, with the government ready, willing, and able to proceed forthwith. But since Denise has no good speedy trial claim-all this delay occurred preindictment, and thus outside her accusation period-Cindy should have no good speedy trial claim either.5 7 Because no speedy trial right is violated in Cindy's and Denise's cases, no issue of speedy trial remedy-of possible dismissal with prejudice-should properly arise. Beyond any applicable statute of limitations protection, Cindy and Denise may also try to invoke the broad innocence-protection principle underlying the Due Process Clause. Unlike the clear words and logic of the Speedy Trial Clause that focus only on the accusation period and the distinctive harms caused by accusation, due process principles can focus on the threat to innocence posed by evidentiary staleness over the entire period between the crime and the trial. (This is the meaning of proposition (5), derived from the Ewell, Marion, and Lovasco cases.)5 8 But in the absence of intentional governmental misconduct-purposefully delaying indictment for mere tactical advantage, in the hope, say, that a key alibi would die-Cindy and Denise are unlikely to win dismissal with prejudice under the Due Process Clause. Their main protection will come from another innocence-protecting due process principle-the Winship principle that the government must prove its case beyond reasonable doubt. 9 Cindy and Denise will be free to argue at trial that the staleness of the case should raise reasonable doubt in the minds of the jury. Because the government bears a heavy burden of proof, added uncertainty caused by long lapses of 55. See supra text accompanying notes 33-34; United States v. Marion, 404 U.S. 307, 313-21 (1971); United States v. MacDonald, 456 U.S. 1, 6-9 (1982); United States v. Loud Hawk, 474 U.S. 302, 310-12 (1986). 56. See Doggett v. United States, 505 U.S. 647, 662-63 (1992) (Thomas, J., dissenting) (asserting that Speedy Trial Clause meant only to protect defendant from oppressive incarceration or anxiety of known criminal charges caused by accusuation, not prejudice to defense caused by passage of time). But see id. at 654-55 (opinion of the Court, per Souter, J.) (asserting that speedy trial inquiry must weigh effect of delay on accused's defense). Cf. supra note 49 (suggesting a better way-between Thomas and Souter-to analyze the problem posed by Doggett). 57. For a very similar analysis, see United States v. Ewell, 383 U.S. 116, 122 (1966). 58. See supra text accompanying note 35; Ewell, 383 U.S. at 122; Marion, 404 U.S. at 322-25; United States v. Lovasco, 431 U.S. 783, 789 (1977). 59. In re Winship, 397 U.S. 358, 362-64 (1970).

656

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

time often helps defendants.6" In the hypotheticals discussed thus far, dismissals with prejudice under proposition (7) would seem to undercut the logic underlying propositions (1)-(6).61 Contrary to proposition (7), dismissal with prejudice is not the "only possible" remedy, or even an apt one in many situations. But is it ever appropriate, as a constitutional mandate? Let's consider again defendants Andy, Bill, and Cindy. Suppose, in Andy's case, that unconstitutionally prolonged pretrial detention--detention beyond the permissible one month period-itself caused irreversible loss of evidence and the like, to Andy's detriment, and fundamentally weakened his ability to present key exculpatory evidence and make his defense at trial. Here, Andy might plausibly claim that any trial would itself be unfair, and that the unfairness was related, both causally and analytically (i.e., both factually and legally), to a constitutionally unspeedy accusation period. He was in jail too long only because he was accused too long, and he has irreversibly lost access to key evidence because of his incarceration, and not merely because of the passage of time.62 In such a case, dismissal with prejudice might be an apt remedy. But note here how this remedy has now been pegged to innocence protection and truth-seeking: Because of the government's own constitutional violation, reliable exculpatory evidence and the like have vanished. Even here, dismissal with prejudice is a severe sanction. A lesser fair trial remedy6 3 might let Andy's lawyers tell the jury his sob story of governmentcreated impediments to his efforts to locate key evidence and witnesses, and let the jury draw whatever inferences it chooses, with Andy benefitting from Winship's command to resolve all reasonable doubt in his favor.' But 60. See United States v. Loud Hawk, 474 U.S. 302, 315 (1986) (holding that possibility of prejudice due to delay not sufficient to support speedy trial claim; passage of time may make it "difficult or impossible" for government to prove case beyond reasonable doubt). 61. What's more, as a practical matter, the extreme nature of the dismissal remedy often leads judges to strain to deny, on the facts of the case at hand, that the Sixth Amendment was indeed violated. These strained denials undermine the values of the clause, as articulated in propositions (1)-(6). See Amsterdam, supra note 38, at 539-43. For a possible example in the case law, see Loud Hawk, 474 U.S. at 317. Twice calling attention to the severity of dismissal, id. at 314, 317, the Court glossed over troubling delay created by the Ninth Circuit's egregious and unexplained footdragging in processing interlocutory appeals. See id. at 324-25 (Marshall, J., dissenting). But see id. at 325 n.8 (noting that, on the facts of the case, defendants perhaps invited delay and thus might deserve to lose). 62. If the damage is not irreversible, then other remedies may undo the damage and make a fair trial possible. Such remedies might include, for example, court-appointed special detectives to help find lost evidence, and continuances at Andy's request to put together those parts of his defense that incarceration impeded. 63. In addition to Andy's independent remedies for illegal detention simpliciter, see supra text accompanying note 45. 64. This might also be the best response to government-created impediments that may not be independently unconstitutional, as when a defendant charged with crime A is already lawfully serving time for a separate crime B, perhaps even in another jurisdiction. (The latter situation also raises nice questions of dual sovereignty.) The wrinkles raised by these

1996]

SIXTH AMENDMENT FIRST PRINCIPLES

657

if instead dismissal with prejudice occurs, 6 5 it occurs precisely to protect a possibly innocent Andy because of hispossible innocence. It's hard to think of a similar scenario for (accused but undetained) Bill. It might be argued that the longer Bill's name is unconstitutionally tarred

by an overly extended accusation, the harder it will be to find a fair jury, and thus trial fairness itself is at stake. But prospective jurors rarely know about indictments in run-of-the-mill cases; are dismissible for cause unless they promise to base their verdict solely on the evidence presented at trial; and are told that accusation itself is not evidence. (That, after all, is one of

the core meanings of the legal presumption of innocence.) What's more, the possibility of continuances and venue transfers make it highly unlikely that any harm to a fair trial caused by Bill's overlong accusation will truly be irreversible.66 (To avoid subjecting Bill to additional stigma during a

continuance caused by an already overlong accusation period, his indictment could be formally dropped during a continuance period, after which a prosecutor would need to seek a new indictment.)6 7 Finally, consider Cindy's case. Suppose that she can show that prosecutors intentionally delayed indictment solely for tactical advantage, in order to impede her efforts to put on a strong defense at trial. Suppose further that as a result of this strategy she has indeed forever lost certain key exculpatory evidence or testimony. If so, due process principles might justify dismissal with prejudice,68 but once again, dismissal would be de-

signed to protect Cindy because of her possible innocence. Dismissal would punish government misbehavior, but on a very different logic than that of the Fourth Amendment exclusionary rule. The exclusionary rule punishes government for trying to introduce reliable evidence, whereas due permutations lie beyond the scope of my analysis today. In the caselaw, see Dickey v. Florida, 398 U.S. 30 (1970); Smith v. Hooey, 393 U.S. 374 (1969). On dual sovereignty generally, see Amar & Marcus, supra note 18, at 4-27. 65. Dismissal could be granted either before or after trial. In unclear cases, holding the trial will enable the judge to better gauge precisely how much the delay has impaired an adequate defense. See United States v. MacDonald, 435 U.S. 850, 858-59 (1978). 66. For another example of curable injury created by overlong accusation, suppose that Bill can show that during the last five months of his accusation period-the five unconstitutional months-he lost his job because of the extra stigma caused by overlong public accusation, and now he can no longer afford his high-priced defense lawyer. Bill has indeed suffered an accusation-based injury that has created a risk of trial unfairness, but dismissal with prejudice is hardly the only possible remedy. Direct payment of Bill's high-priced lawyer by the government itself would-for fair trial purposes-put Bill in the same position he would have been in had the government dropped the indictment for the last five months, or tried Bill five months earlier. 67. See supra notes 49, 53 and accompanying text; infra note 80. 68. See United States v. Lovasco, 431 U.S. 783, 795-96 (1977) (holding that investigative delay, as opposed to bad-faith tactical delay, not due process violation even if defense somewhat prejudiced by time lapse); United States v. Marion, 404 U.S. 307, 324 (1971) (holding that Due Process Clause might require dismissal of indictment if shown at trial that preindictment delay caused substantial prejudice to fair trial rights and delay was intentional device to gain tactical advantage).

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

process dismissal with prejudice would punish government for, in effect, trying to suppress a defendant's reliable evidence.6 9

The foregoing analysis has been based on general remedial logic, in the light of propositions (1)-(6) in current caselaw. Several big ideas about the

Speedy Trial Clause have done much of the work. For starters, the clause focuses distinctively and exclusively on harms created during and by criminal accusation. Next, these harms take three distinct shapes, implicating physical liberty, reputation and peace of mind, and reliable trials. Finally, dismissal with prejudice is never an apt response to physical liberty and reputation harms. Rather, dismissal is appropriate only in cases of reliable trial harm-in cases where the trial itself poses an unacceptable risk that,

because of the government's own prolongation of accusation, an innocent man may be erroneously convicted.

What remains is to root these big ideas about Speedy Trial Clause rights and remedies in constitutional text, history, and structure. B. TEXT, HISTORY, AND STRUCTURE

1. Rights The Sixth Amendment proclaims that "[in all criminal prosecutions, the accused shall enjoy the right to a speedy ... trial." By its terms, this clause seems to apply only when the government has "accused" a person and initiated a "criminal prosecutionf]." Elsewhere the Amendment's District Clause speaks of the time when "the crime shall have been committed,"

but contains no language suggesting that this is the moment when the speedy trial clock starts ticking. On the contrary, the tense of the District Clause's "shall have been committed" plainly contrasts with the tense of the Speedy Trial Clause's "shall enjoy," suggesting that the speedy trial right comes into play after the crime, not alongside it. Indeed, the Amend69. It is possible to imagine an interesting variant of due process dismissal in the context of an outrageous search and seizure, a variant that might be seen as an "inverse exclusionary rule." Suppose the government is planning to introduce Exhibit A against Cindy, but knows that Cindy will counter with Item B, which will dramatically undercut or neutralize A's impact. Suppose that the cops illegally and outrageously break into Cindy's house, find B, seize it, and destroy it. Beyond Cindy's obvious compensatory and punitive Fourth Amendment tort remedies, Cindy has a great due process, fair-trial, innocence-protection argument that the government's Exhibit A should be excluded from the trial. But this inverse exclusionary rule differs from its standard Fourth Amendment evil twin in key ways. It excludes Exhibit A, rather than Item B-the thing actually seized. Most important, it excludes A because government conduct has rendered A presumptively unreliable (if A was really so accurate, why did the government destroy B?); the standard exclusionary rule excludes evidence of the highest reliability and probative value. Inverse exclusion is designed to protect the innocent because of their innocence; standard exclusion is designed to protect the guilty defendant as such. For a case whose logic might support inverse exclusion as a response to malicious destruction of evidence, see Arizona v. Youngblood, 488 U.S. 51, 56-59 (1988) (finding no due process violation in absence of bad-faith destruction of evidence).

1996]

SIXTH AMENDMENT FIRST PRINCIPLES

ment, in its entirety, proclaims that "[i]n all criminal prosecutions, the accused shall enjoy the right ... to be informed of the nature and cause of the accusation." This modified repetition of the word "accused" strongly confirms that the Sixth Amendment as a whole is accusation-based. It is hard to imagine what the "nature and cause" clause could possibly mean prior to governmental accusation; and since this clause is syntactically interwoven with the very word "accused" in the Speedy Trial Clause, plain

grammar reinforces plain meaning: "accused" must mean "accused." Common sense provides further support: prior to accusation, how can we know precisely which statutorily defined crime the government must prove at a speedy trial? As the Court put the point in Marion, "[t]he framers could hardly have selected less appropriate language if they had intended the 70 speedy trial provision to protect against pre-accusation delay." History and structure confirm the point. The Marion Court found "nothing in the circumstances surrounding the adoption of the Amendment indicating that it does not mean what it appears to say."7 1 On the contrary, at the time of the Founding, leading common law commentators and landmark English legislation addressing trial timing had identified pretrial

detention-detention triggered by governmental accusation rather than the commission of the crime simpliciter-as a key concern.7 2 The Eighth

Amendment's prohibition of "[e]xcessive bail" reflects a similar concern, thus reinforcing our reading of the Sixth. 73 Even more dramatic reinforcement for an accusation-based reading of the Sixth Amendment comes from the Fifth Amendment's requirement of grand jury indictment for any serious federal crime.7 4 Precisely because criminal accusation itse!fcreates distinctive risks to the "accused," the accusation process requires special

70. Marion, 404 U.S. at 314-15. A contrary argument would concede that of course the Amendment in one sense does not apply unless and until one becomes "accused": an

unaccused person has no right to demand immediate arrest or indictment. But once one becomes "accused," the argument goes, one can retroactively insist that the speedy trial clock started ticking when the crime occurred. This is textually possible, but strained. No other clause in the Amendment sensibly applies "retroactively," and there is next to no historical support in the Founding era for this textual and temporal somersault. 71. Id. at 313-14. 72. See EDWARD COKE, THE SECOND PART OF THE INSTITUTES OF THE LAWS OF ENGLAND

43 (Brooke, 5th ed. 1797) ("[Justices] have not suffered the prisoner to be long detained, but at their next comming have given the prisoner full and speedy justice, . . . without detaining him long in prison.") (emphasis added); Habeas Corpus Act of 1679, 31 Car. 2, ch. 2. For more on the act, see infra text accompanying notes 83-85. 73. U.S. CONST. amend. VIII ("Excessive bail shall not be required. ). The protections of the Eighth Amendment, of course, in no way eliminate the need for the independent safeguards of the Sixth. Not all offenses are bailable-several capital crimes, for example, were historically not subject to bail. See, e.g., 4 BLACKSTONE, supra note 52, at *298-99. Even nonexcessive bail might be set at high levels in the event of a high objective risk of flight, and many defendants might not be able to post such high bail. 74. See U.S. CONST. amend. V ("No person shall be held to answer for a capital, or otherwise infamous crime, unless on a presentment or indictment of a Grand Jury.").

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

safeguards; no person will be forced to run the risks of accusation unless a large panel representing the people-a grand jury-has decided to subject him to those risks. All this helps confirm our logical inference that the Speedy Trial Clause protects a person not merely after he has become "accused," but because he has become "accused"-protects him, that is, from the distinctive risks of accusation itself. What are those risks? Here too, a close examination of the Fifth Amendment can help illuminate the Sixth: "No person shall be held to answer for a capital, or otherwise infamous crime, unless on a presentment or indictment of a Grand Jury." Note the word "held." The word is not merely metaphoric but also literal. Upon criminal accusation one's physical body is vulnerable to physical detention, pretrial. Note also the word "person": "No person shall be held .... " This embodied reference immediately follows two embodied references in the Fourth Amendment, protecting Americans' physical "persons" from being "searched" or "seized" unreasonably,7 5 and immediately precedes a dramatically physical reference in Double Jeopardy Clause to the "life or limb" of the Fifth Amendment's "any person.",76 (Together, these four tightly clustered references are the only times the words "persons" and "person" appear in the Bill of Rights.) Precisely because one's very body may be held upon accusation of criminal wrongdoing, the Speedy Trial Clause demands that a trial occur speedily after accusation. At such a trial, of course, an innocent man can, in the words of the Fifth Amendment, "answer" his "indictment"-can, in the words of the Sixth Amendment, make "his defence" against the "criminal ...accusation." And if the innocent man prevails at this speedy, public, and fair trial, he immediately gets his "person" back and wins release from the cell in which he may be "held" during the accusation period. But public accusation threatens more than a person's body; it also assaults his good name. The text of the Speedy Trial Clause, after all, is

75. U.S. CONST. amend. IV (affirming Americans' right to "be secure in their persons,

houses, papers, and effects, against unreasonable searches and seizures" and mandating that warrants particularly describe "the persons or things to be seized"). 76. U.S. CONST. amend. V ("[N]or shall any person be subject for the same offence to be ").A critic of plain meaning textualism in twice put in jeopardy of life or limb .... constitutional criminal procedure might try to make hay of this Double Jeopardy phrase. Surely, the argument goes, the clause must apply to all serious criminal charges, even if death and dismemberment are not authorized punishments. I agree that a strict expressio unius reading of "life or limb" would be obtuse. Cf. supra text accompanying note 21. But I also suggest that a plain meaning approach to the clause would read the phrase as a term of art, a grimly poetic synecdoche for all serious punishment. There is, I submit, a big difference between plain meaning textualism and tin ear textualism. See Er parte Lange, 85 U.S. (18 Wall.) 163, 170-73 (1873) (reading phrase to encompass all punishment, even misdemeanors); People v. Goodwin, 18 Johns. 187, 201 (N.Y. Sup. Ct. 1820) (reading phrase as metaphor for felonies); JOEL P. BISHOP, COMMENTARIES ON THE CRIMINAL LAw 543 (1856) (similar). But see United States v. Gibert, 25 F. Cas. 1287 (C.C.D. Mass. 1834) (No. 15,204) (Story, Circuit J.) (offering narrow reading).

1996]

SIXTH AMENDMENT FIRST PRINCIPLES

not limited to those "accused" "persons" "held" in pretrial detention, but rather extends to "accused" individuals in "all criminal prosecutions." ' So, to make full sense of the text, we must focus on the harms inherent in every criminal accusation. Every criminal accusation, of course, is an attack on the accused's reputation-a charge of "criminal" wrongdoing in the words of the Speedy Trial Clause, of "infamous" misconduct in the analogous words of the Grand Jury Clause. At common law, a false accusation of criminal behavior was viewed as defamation per se.7 8 (Note the obvious etymological link between defamation and accusations of infamous crime.) And as with pretrial detention, each additional day of accusation was a fresh assault, a new injury. Here too, the innocent man would want a trial to speedily follow accusation, so that he could offer his "answer" and put on his "defence." And if the innocent man can prevail at this speedy, public, and fair trial, he puts an end to the accusation of infamy and wins back his good name. Of course, the innocent man might also lose at trial. If so, the detention (if any) of his body and the necessary assaults on his soul during the accusation period may merely be a foretaste of the injuries and indignities that may be heaped upon him after conviction. And here we see the final injury inherent in accusation itself: accusation puts a person on a path of peril. Formally, accusation and indictment shove a person onto a legal road whose destination can never be certain even for the innocent man. At the end of this road may hang a noose, as the Fifth Amendment reminds us in no less than three grim phrases: "capital... crime," "jeopardy of life or limb," and "deprived of life, liberty or, property."7 9 As we shall see, much of the rest of the Sixth Amendment was designed to reduce the risk that a noose would be wrapped around an innocent neck; and so it would be structurally odd if the Speedy Trial Clause of that same Amendment ignored this risk. On the contrary, it makes sense to read the clause as prohibiting situations where an extended accusation period itself could substantially increase the likelihood of an innocent man being erroneously convicted. Again, the risk to innocence under the clause must not come from the mere passage of time-the standard evidentiary staleness issue addressed by statutes of limitations-but must somehow be traceable to an extended accusation.80 77. U.S. CONST. amend. VI (emphasis added). 78. See, e.g., 3 BLACKSTONE, supra note 52, at *123. 79. U.S. CONST. amend. V. 80. Three situations suggest themselves. First, the clause must prevent extended pretrial detention that would itself materially impede the incarcerated defendant's ability to assemble his defense. See supra text accompanying notes 62-65. Likewise, the clause must prevent extended accusations that may ostracize and impoverish the accused in civil society, and thus undermine his ability to pay for his defense. See supra note 66. Finally, the clause might plausibly be read to demand that after a certain time period, a grand jury indictment necessarily lapses. See supra note 49. This would ensure that the two public verdicts

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

To insist that the Speedy Trial Clause focuses not on ordinary evidentiary staleness, but only on threats after and because of accusation, does not leave defendants in egregiously stale cases constitutionally naked. The broad language of the Due Process Clause is not limited by the language and logic of accusation; and so it can properly be deployed to address long delays between crime and trial.8 ' The Due Process Clause must supplement the Speedy Trial Clause in at least one other respect. Although a

defendant may seek to waive his right to a speedy trial, the Speedy Trial Clause does not, by its terms, confer on him a general constitutional right to an unspeedy trial. In general, a defendant has no constitutional right to stop the trial train (perhaps with the hope that governmental evidence will,

over time, fade faster than his own defense evidence). But if the government seeks to go to trial so quickly that a defendant truly cannot assemble his defense, at some point due process will demand delay, lest an innocent

man suffer an undue risk of erroneous conviction. Though this due process right goes beyond the text of the Speedy Trial Clause, its innocenceprotecting logic vindicates the deep structure of that clause, of the Sixth as a whole, and of constitutional criminal procedure generAmendment 82 ally. 2. Remedies Having consulted text, history, and structure to deduce the proper nature of speedy trial rights, we must now look to the same sources for guidance on speedy trial remedies. rendered by two different panels of the people-the grand jury and the petit jury-are roughly contemporaneous in time. The textual proximity of the Fifth Amendment Grand Jury Clause and the Sixth Amendment Petit Jury Clause, with the Speedy Trial Clause somewhere in between, is perhaps suggestive of a desired temporal proximity of the two juries' verdicts. Temporal proximity would help safeguard innocence by limiting the ability of prosecutors to forum shop over time. (Imagine a prosecutor who cajoles a single, unusually proprosecutor grand jury to issue stacks of indictments, and then stockpiles these-perhaps without even making them public, to avoid triggering any reputation interest-until unusually proprosecutor petit venires materialize from time to time.) 81. In many cases, however, long delay may be wholly justifiable. The crime may not come to light for many years; or the government, despite due diligence, may not have sufficient evidence early on to warrant prosecution; or prosecutors may hold back to avoid compromising ongoing investigations; or .... In many of these cases, defendants may gain more from delay than they lose, given that evidentiary staleness can create reasonable doubts that must, under Winship, be resolved against the government. See United States v. Loud Hawk, 474 U.S. 302, 315 (1986). A constitutional superstatute of limitations rooted in due process should bar prosecutions only in cases where prosecutors manipulate timing solely for tactical advantage or to vex defendants, and leave the rest to Winship. 82. In the Speedy Trial Clause caselaw, see Barker v. Wingo, 407 U.S. 514, 521 & n.15 (1972); United States v. Ewell, 383 U.S. 116, 120 (1966). See also Powell v. Alabama, 287 U.S. 45, 71 (1932) (holding that "failure of the trial court to give [defendants] reasonable time and opportunity to secure counsel was a clear denial of due process"); Peter Westen, The Compulsory Process Clause, 73 MICH. L. REV. 71, 109 (1974) (arguing that trying a defendant "before he can call witnesses violates the right of compulsory process").

1996]

SIxTH AMENDMENT FIRST PRINCIPLES

663

a. The Bodily Liberty Interest. Consider pretrial detention. The Founders' first line of defense came from the glory of the common law, the Great Writ of habeas corpus. Judges must issue this writ to end or soften the conditions of unduly oppressive pretrial detention. The aptness of the Great Writ to address the specific problem of pretrial detention was the centerpiece of the celebrated English Habeas Corpus Act of 1679.83 This document stood alongside Magna Charta and the English Bill of Rights of 1689 as a towering common law lighthouse of liberty-a beacon by which framing lawyers in America consciously steered their course. 84 We should not then be surprised that the only provision in the entire Constitution addressing the technical issue of remedies with any specificity was an 85 Article I clause safeguarding the Great Writ. But even though habeas must be the first line of defense, it cannot be the only gun in our remedial arsenal. If, in Andy's case, a judge properly issues a writ on day 30, ordering Andy's release, habeas here is less a remedy than a prevention of any violation from ever occurring. 6 If, instead, a judge issues the writ only weeks or months later, habeas can limit the violation of bodily liberty and prevent fresh assaults, but it does not remedy the wrong done by detention after day 30 but before the writ issues. Habeas does not make Andy whole. Here is where after-the-fact compensatory and punitive damages come in. Consider, for example, the Framers' clearly established regime for vindicating the Fourth Amendment-for remedying past wrongs and deterring future wrongs. Suppose a constable in the early Republic unconstitutionally arrested John Doe, without a warrant, without good reason, but with, say, a statute that conferred sweeping and arbitrary arrest powers on all constables. This arrest was, of course, a trespass against Doe's person. Doe could sue the constable in trespass, seeking compensatory and punitive damages for the outrage upon his person. The constable, in turn,

83. Habeas Corpus Act of 1679, 31 Car. 2, ch. 2. 84. See Amar, supra note 53, at 1205. 85. U.S. CONST. art. I, § 9, cl. 2 ("The privilege of the Writ of Habeas Corpus shall not be suspended, unless when in Cases of Rebellion or Invasion the public Safety may require it."). On the uniqueness of this clause as an explicit statement about remedies in the Constitution, see Alfred Hill, ConstitutionalRemedies, 69 COLUM. L. REV. 1109, 1118 n.42 (1969). 86. A complementary and more modern scheme of prevention-an administrative law "translation" of the common law habeas scheme-would feature a regulatory framework statute like the federal Speedy Trial Act, 18 U.S.C. §§ 3161-74 (1994), laying down parameters for reasonable case management. For a discussion of the importance of such regulatory regimes, see John C. Godbold, Speedy Trial-Major Surgery For a National Ill, 24 ALA. L.

REV. 265 (1972). Cf. Amar, Fourth Amendment, supra note 2, at 816-17 (discussing need for similar legislative and administrative regimes in Fourth Amendment context). Though my focus today is on constitutional doctrine, this focus must not obscure the hugely important role that framework statutes, like the Speedy Trial Act, have played and must continue to play.

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

would plead governmental authority: a statute authorized the trespass and in effect trumped the tort suit. In a world without a written Constitution, the constable's answer might have been a winner. Government (the argument would go) may do things that private persons may not; and so if the government authorizes its agent to pick your pocket, it's not called trespass-it's called taxes. But the Framers gave Americans a world with a written Constitution, and in Doe's case this would make all the difference in the world. In response to the constable's invocation of lawful governmental authority, Doe would deny that the authority was indeed lawful. If the statute sought to authorize unreasonable seizures of persons-without good cause, without standards-the statute was itself unconstitutional under the Fourth Amendment. It was null and void-ultra vires-and thus no good defense. If a court agreed with Doe that the statute was unconstitutional under the Fourth Amendment, the constable's defense would fall, and Doe could win compensatory and (in the case 6f egregious conduct) punitive damages. And this was so even if the constable had acted in the good faith, but erroneous, belief that the authorizing statute was wholly constitutional. As I have shown elsewhere, this was the clear Founding paradigm for 87 vindicating the Fourth Amendment. More recent trends have updated this paradigm. In the landmark 1971 Bivens case, for example, the citizen was allowed to sue abusive federal officials for compensatory and punitive without the need to damages directly under the Fourth Amendment itself, 88 trespass. law common predicate a plead and prove This clear paradigm of Fourth Amendment remedies has obvious implications for Sixth Amendment remedies. How is Andy any different from John Doe? How is a jailer unconstitutionally holding Andy in violation of the Sixth Amendment any different from a constable unconstitutionally seizing Doe in violation of the Fourth Amendment? If Doe can sue in trespass and win against the constable under the Fourth, why can't Andy sue in trespass and win against the jailer under the Sixth? If, today, a Bivens suit for damages lies directly under the Fourth, why not under the Sixth too? Indeed, we can sharpen the point even finer. Isn't Andy's detention, if unconstitutional under the Sixth, also necessarily a Fourth Amendment violation too? The Fourth Amendment, of course, prohibits all "unreasonable ... seizures [of] persons." Unreasonableness here surely encompasses 89 seizures of persons for unreasonably long time periods. And surely a 87. See Amar, Fourth Amendment, supra note 2, at 771-79, 812-15. 88. Bivens v. Six Unknown Named Agents of the Fed. Bureau of Narcotics, 403 U.S. 388, 390-97 (1971). 89. See United States v. Place, 462 U.S. 696, 707-10 (1983); Money v. Leach, 19 Howell's State Trials 1001, 1026 (K.B. 1765) (Mansfield, C.J.); Amar, Fourth Amendment, supra note 2, at 776, 807.

1996]

SIXTH AMENDMENT FIRST PRINCIPLES

seizure of a person that violates another provision of the Constitutionhere, the Sixth Amendment-must be viewed as constitutionally unreasonable within the meaning of the Fourth Amendment. 90 At first, treating Sixth Amendment violations like Fourth Amendment violations might seem to support dismissal with prejudice by analogy to the Fourth Amendment exclusionary rule. But we are seeking here principles from constitutional text, history, and structure, and none of these supports the so-called Fourth Amendment exclusionary rule. The text of the Fourth Amendment says nothing about criminal exclusion; indeed, it nowhere distinguishes between government-initiated criminal cases (where exclusion now holds sway) and government-initiated civil cases (where exclusion is not now and never has been the rule). The Fourth Amendment's text, by contrast, does clearly presuppose tort law and property law-trespass rules and the like-protecting Americans in their "persons, houses, papers, and effects." No English court at the founding had ever excluded reliable evidence on proto-Fourth Amendment grounds; nor has any English court since. Though most state constitutions featured state counterparts to the federal Fourth Amendment, no court in America-state or federal-ever excluded evidence on search-or-seizure grounds prior to 1886. By contrast, both English courts in the 1760s and early American courts, in landmark cases known to all lovers of liberty, awarded liberal tort remedies against abusive government searchers and seizers. 91 When exclusion finally came to America in 1886, it came not as a remedy for a past Fourth Amendment wrong, but rather as a kind of prevention of a threatened Fifth Amendment harm: to introduce as evidence, in a criminal case, a man's illegally obtained private papers would be tantamount to making him an involuntary witness against himself in a criminal case.9 2 (And now we can see why exclusion-on Fifth Amendment groundsapplied only in criminal but not civil cases.) By a gradual and not always self-conscious process of extension, this rule for illegally seized diaries and private papers came to be applied to a great deal of other illegally seized "fruit." 9 3 As I have explained in detail elsewhere, this fusion of the Fourth and Fifth Amendments, though intriguing, was wrong from the outset 94

90. See Amar, FourthAmendment, supra note 2, at 804-05.

91. For much more documentation of the claims summarized in this paragraph, see generally Amar, Fourth Amendment, supra note 2.

92. See Boyd v. United States, 116 U.S. 616, 634-35 (1886). 93. See, e.g., Weeks v. United States, 232 U.S. 383 (1914); Silverthorne Lumber Co. v. United States, 251 U.S. 385 (1920); Gouled v. United States, 255 U.S. 298 (1921); Amos v. United States, 255 U.S. 313 (1921); Agnello v. United States, 269 U.S. 20 (1925). See

generally Osmond K. Fraenkel, Recent Developments in the Law of Search and Seizure, 13

MINN. L. REV. 1, 4-5 & nn.48-52 (1928) (citing cases from five states declining to apply Weeks rule to contraband). 94. See generally Amar, Fourth Amendment, supra note 2, at 785-800; Amar & Lettow, Fifth

Amendment, supra note 3.

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

95 and has now been decisively rejected by the Supreme Court. Without it, the so-called exclusionary rule has no text, history, or structure to stand on. What's more, even at the height of the exclusionary rule, an exception for unconstitutional seizures of persons was always recognized. Even if the government kidnapped a suspect in violation of every norm of civilized conduct and every Fourth Amendment principle in the book, it could 1952 Court, per nonetheless hold him for criminal trial. As a unanimous 96 Collins: v. Frisbie in Justice Black, put the point

This Court has never departed from the rule announced in [1886] that the power of a court to try a person for crime is not impaired by the fact that he had been brought within the court's jurisdiction by reason of a "forcible abduction." ... There is nothing in the Constitution that requires a court to permit a guilty person rightfully convicted to escape justice because he was brought to trial against his will. If we substitute "after an overlong accusation period" for "by reason of a 'forcible abduction' " and "against his will" in this quote, we can see strong 97 precedential support, even within the exclusionary rule tradition, for rejecting Sixth Amendment dismissal with prejudice. Perhaps the implicit logic of Frisbie was that holding and using the defendant's body itself would not be testimonial witnessing in violation of the Fifth Amendment; or perhaps the Court simply saw the particularly dramatic upside-down effect that a judicial release order might entail, and balked at the root idea of exclusion. (Release would be a huge windfall to the guilty, but would fail to make the innocent kidnap victim whole.) In any event, this caselaw confirms the need to develop tort remedies to vindicate the rights of innocent detainees. And so we return to the Framers' remedial paradigm that, as a general matter, held constables liable. If constables, why not jailers? 95. See United States v. Leon, 468 U.S. 897, 905-06 (1984); Fisher v. United States, 425 U.S. 391, 407 (1976).

96. 342 U.S. 519, 522 (1952). Though decided prior to the era of general "incorporation" of the Bill of Rights against the states, Frisbie was authored by the Court's leading proponent of total incorporation and contains no language suggesting that federal kidnapping would somehow be different from the state kidnapping at issue in Frisbie. 97. See also INS v. Lopez-Mendoza, 468 U.S. 1032, 1039 (1984) ("The 'body'... of a defendant ... is never itself suppressible as a fruit of an unlawful arrest."); United States v. Crews, 445 U.S. 463, 474 (1980) (opinion of the Court, per Brennan, J.) ("An illegal arrest, without more, has never been viewed as a bar to subsequent prosecution, nor as a defense to a valid conviction. ... Respondent is not himself a suppressible 'fruit.' "); Gerstein v. Pugh, 420 U.S. 103, 119 (1975) (reiterating "established rule that illegal arrest or detention does not void a subsequent conviction"). All of these case cite Frisbie. Cf. United States v. Blue, 384 U.S. 251, 255 (1966) ("[The exclusionary rule] does not extend to barring the prosecution altogether. So drastic a step ... would also increase to an intolerable degree interference with the public interest in having the guilty brought to book.").

1996]

SIXTH AMENDMENT FIRST PRINCIPLES

Our jailer, however, has an ace up his sleeve. Whereas our constable could point only to an unconstitutional statute, our jailer can point, in effect, to a court order authorizing Andy's pretrial detention. This court order is, by hypothesis, unconstitutional, but was issued by a court with jurisdiction. And at common law (crafted by judges) court orders (issued by judges) had special powers: an officer carrying out a substantively incorrect, but jurisdictionallyauthorized, court order could hide behind the order and escape liability. In the Fourth Amendment context, for example, a constable who executed a warrant that a judge had wrongly-indeed, unconstitutionally-issued (say, a warrant unsupported by probable cause, or authorizing an objectively unreasonable search) could hide behind the warrant.98 Our jailer would argue for like immunity. Nor could Andy sue the judge who ordered this overlong detention because, once again, the judge's actions, even though substantively unconstitutional, were jurisdictionally authorized. The typical remedy for an incorrect judicial act is an appeal to a higher court, but this generally satisfactory remedy rings hollow in a few atypical situations where the real constitutional injury-here, day upon extra day of unconstitutional detention-is inflicted while an appeal is pending.9 9 In a world without a written Constitution, this remedial shell gamejailers hiding behind judges, judges hiding behind immunity and appeals, and appeals coming too late to vindicate the real legal interest at stakewould be bad enough; but in a world with a written Constitution it makes even less sense. The Constitution declares rights directly against the government itself. This is the lesson of Bivens;1 00 and so when government, through its agents-judicial or executive or some combination-violates those rights, the government itself should be held liable for damages. Rights against government itself should be vindicated by remedies against government itself. As I have argued elsewhere, when government violates the express limits on its powers imposed by We the sovereign People in our Constitution, government, properly speaking, is neither "sovereign" nor "immune" and cannot in justice or logic invoke "sovereign immunity. 10°1 And so, by analogy to and extension of constitutionally proper Fourth Amendment remedies for illegal seizures, we can deduce constitutionally proper Sixth Amendment remedies for illegal detention: Andy should be allowed to sue the detaining government for compensatory and (in cases of

98. See Amar, FourthAmendment, supra note 2, at 781. 99. On the unsuitability of mandamus to solve all speedy trial problems, see Note,

Dismissal of the Indictment as a Remedy for Denial of the Right to Speedy Trial, 64 YALE L.J. 1208, 1209 n.9 (1955). 100. Bivens v. Six Unknown Named Agents of the Fed. Bureau of Narcotics, 403 U.S. 388, 390-97 (1971). 101. See Akhil Reed Amar, Of Sovereignty and Federalism, 96 YALE L.J. 1425, 1466-92

(1987).

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

intentional or egregious violations) punitive damages. At the Founding, suits against the government itself were recognized de facto: a citizen would sue an officer in his personal capacity, but everyone understood that the real party in interest was the government itself, which would typically be forced to indemnify officials who were merely carrying out government policy. 1°2 (Without indemnification for good faith service, who would agree to work for the government?) Precisely because officials would be indemnified, it was not unfair to hold them strictly liable for constitutional torts, even if they acted on the good faith (but erroneous!) belief that their behavior was fully constitutional. Thus, the Framers did not recognize, and would have rejected, the twentieth-century notion of "good faith immunity." In other words, many Framers, well before Coase, understood the against Coase Theorem-understood, that is, that tort suits nominally 03 individual officers were in reality suits against "the public purse."' Once we see the logic of the Framers' ingenious remedial system-de facto government strict liability for constitutional torts-we can see how the system fails in certain unusual cases. If a series of governmental actors together violate a citizen's constitutional rights, but no one of them commits an actionable common law tort, the citizen has no one to sue, and the Framers' ingenious system of common law remedies fails its° central purpose: to assure full remedies for every constitutional right." And it was precisely this recognition that properly led the Court in Bivens to infer a damage action directly under the Constitution in a case where a constitutional violation had occurred without a predicate common law violation. But the Bivens Court stopped half-way and recognized a cause of action only against officers and not the government itself. If a right derives directly from the Constitution, and runs directly against the government itself, then the damage remedy should run directly against the government 0 5 for every right against government itself. Following Marbury's logic,' there should be a remedy against government, de facto or de jure. And the 102. See Amar, FourthAmendment, supra note 2, at 812-13.

103. See id. 104. For a general discussion of this remedial vision, see Amar, supra note 101, at 1484-95. 105. See Marbury v. Madison, 5 U.S. (1 Cranch) 137, 163 (1803) (quoting 3 BLACKSTONE, supra note 52, at *23, *109): The very essence of civil liberty certainly consists in the right of every individual to claim the protection of the laws, whenever he receives an injury. One of the first duties of government is to afford that protection.... "[it is a general and indisputable rule, that where there is a legal right, there is also a legal remedy by suit or action at law, whenever that right is invaded." ... "[E]very right, when withheld, must have a remedy, and every injury its proper redress."

The government of the United States has been emphatically termed a government of laws, and not of men. It will certainly cease to deserve this high appellation, if the laws furnish no remedy for the violation of a vested legal right.

19961

SixTH AMENDMENT FIRST PRINCIPLES

Sixth Amendment speedy trial right is most definitely a right against government as such. Only the government, after all, can conduct a trial, speedy or no; and private persons are not generally allowed to "hold" others against their will, unless they have been so authorized by the government. A proper Bivens-like scheme to remedy unconstitutional pretrial detention would dramatically avoid the upside-down effect of dismissal with prejudice. Under an apt scheme a guilty Andy would never recover more than an innocent Andrea, and might sometimes recover less. Suppose that Andy and Andrea have each suffered six months of pretrial detention, the last five of which were constitutionally unspeedy. Each was entitled to habeas release on day 30, but wrongly denied it. Between them, the judge and the jailer-both agents of the state-violated Andy's and Andrea's constitutional rights, even if neither official committed an actionable common law wrong. If Andrea's criminal case goes to trial and Andrea wins acquittal, let's say that she should get $50,000 in her later Bivens-like suit against the government. 10 6 But suppose that Andy, unlike Andrea, is later brought to trial and convicted, and receives a ten year sentence, with a six-month set-off for pretrial time already served, and nine and one-half years to go.' °7 If Andy brings a Bivens-like suit against the government, is he, too, entitled to $50,000-or, indeed, to anything? Admittedly, he wrongly served five months pretrial; but had he not been so detained, he would be serving five extra months post-trial. In terms of his libertydeprivation interest, the unconstitutionality does not appear to have caused any incremental deprivation of liberty, but only accelerated the deprivation. And so here, perhaps, we have a kind of harmless error, expost. Tricky problems arise in intermediate cases. Suppose Andy is criminally tried, convicted, sentenced to six months, and released on the basis of time already served. At face value this, too, is an arguable case of harmless error, ex post; Andy's unconstitutional pretrial detention accelerated his loss of liberty, but did not add to it. But suppose the six-month sentence was a judicial sham; had Andy not already served six months pretrial, he

106. This Bivens-like remedial regime might sensibly borrow from administrative law as well as the common law. Cf.supra note 86. (For a similar suggestion of administrative'law remedies for Fourth Amendment violations, see Amar, Fourth Amendment, supra note 2, at 816.) A workers' compensation-like formula for determining damages could significantly lower adjudicatory transaction costs. Note also that further transactional economies exist because: (1) defendant and the government are already parties before a court that (2) is already supposed to be attending to the effect of time on the case and (3) the defendant already has legal counsel (by court appointment in a case of indigency). Thus, instead of pursuing Bivens-like remedies in collateral civil proceedings, perhaps defendants should be allowed to bring their claims in a kind of contempt proceeding pendent to the criminal prosecution itself. 107. See, e.g., 18 U.S.C. § 3585(b) (1994) (authorizing set-offs for pretrial detention in federal cases).

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

10 8 would have simply been released after conviction. If so, unconstitutional pretrial detention really did rob him of months that he would never have lost, and his case looks more like Andrea's. Bivens-like courts must thus be alert to shams, but however they decide these cases, one thing is sure: a guilty Andy should never recover more than an innocent Andrea, and might sometimes justly recover less. And the reason for this is that the deep logic of the Sixth Amendment, and of constitutional criminal procedure generally, is to protect innocence.

b. The Reputation Interest. Similar principles apply to the reputation interest. As with habeas in the context of bodily liberty, the Constitution's first line of defense is prevention: judges must simply quash indictments that linger too long and thus stain the "accused['s]" reputation without giving him his "speedy ... trial" to "answer" the "infamous' ''indictment" of his character and "defen[d]" against the charge of "criminal" conduct. If quashing occurs in time-before one year in Bill's case-no violation of the reputation right will ever have occurred. If, however, judges fail to quash in time, quashing only prevents new assaults on reputation without making the accused whole. And here, once again, is where after-the-fact, Bivens-like tort damages against the government should come in. If Bill suffers seventeen months of pretrial mud on his name, when the Constitution allows only twelve, Bill's constitutional rights have been violated by the government itself. Obviously Bill cannot sue the grand jurors who indicted him-they, like judges, were acting in a judicial capacity, even if they acted erroneously. Besides, Bill's cognizable injury is not the indictment itself, but the delay in giving him his day in court to clear his name. Nor could Bill, at common law, successfully sue the prosecutor or the judge; the indictment speaks in the name of the government, not the prosecutor personally, and the judge who wrongly delays trial is nevertheless acting within his jurisdiction. But even if no one person has violated Bill's common law rights, together various governmental agents have combined to violate his constitutional rights. Here too, a right against the government itself calls for a like remedy. And here too, an innocent person should never recover less, and, indeed, should typically recover more than a guilty one. If Billy Jo, after seventeen months of accusation, the last five of which were unconstitutional, goes to trial, makes her defense, and wins, she should be compensated for the five extra months of stigma she endured unconstitutionally. (She is not, however, entitled to recover for the first year of accusation, since this was lawful. Similarly, she is not constitutionally entitled to 108. See Amsterdam, supra note 38, at 535-36 n.81 (arguing that trial judges may inflate sentence to offset reductions for pretrial confinement). But see Recent Case, 108 U. PA. L. REV. 414, 422 n.59 (1960) (authored by Anthony Amsterdam) (arguing that sentence reduction may be appropriate remedy).

1996]

SIXTH AMENDMENT FIRST PRINCIPLES

reimbursement for all out-of-pocket legal fees if she wins: indictment and trial themselves are not unconstitutional, even though they inflict great harm.) By contrast, consider the case of guilty Bill, who likewise suffered seventeen months of accusation, but who is convicted at trial. The delay of his trial did heap five months of unconstitutional mud on his name, but had the trial occurred five months earlier, Bill would have been convicted earlier, with even more mud on his name. And so, here too, we see a kind of constitutional harmless error, expost. °9 A similar analysis applies if we substitute a pretrial "anxiety" or "peace of mind" interest for the reputational interest we have been studying. Strong support for this right-side-up remedial effect-where the innocent sometimes benefit more than the guilty-comes from one sensible corner of current Fourth Amendment law. In the 1983 Place case and the 1984 Jacobsen case, the Court dealt with certain governmental intrusions that, according to the Court, could detect only the presence of contraband but could not reveal any other private information."' These intrusions, said the Court, might make criminals nervous, but did not compromise any legitimate interest in privacy. Though the Court's particular approachlabelling these searches nonsearches"1 -was not particularly helpful," 2 its deep instinct was sound: the searches were reasonable, and thus constitutional. And the reason these searches were reasonable is that, although they could ruin a drug runner's day, they posed little threat to the privacy interests of law-abiding folk. Lawbreakers as such have no legitimate interest in privacy, and are at times entitled to less peace of mind than are the law-abiding.113 c. The Reliable Trial Interest. Consider, finally, the reliable trial interest underlying the Speedy Trial Clause: an extended accusation period must not create an undue risk that the trial itself will result in the erroneous conviction of an innocent man. Once again, the first line of defense, by analogy to habeas in the physical detention context, should be prevention. Indeed habeas itself is the first 109. If an acquitted Billy Jo barely wins acquittal-with a trial that suggests that she is probably guilty, but not beyond reasonable doubt-she may receive lower damages than if she wins acquittal by a mile. Analytically, her damages should be measured by the difference in her reputation pre- and post-trial, and by the length of the unconstitutional delay. The more innocent the trial shows her to be, the more she wrongly suffered by unconstitutional trial delay. 110. United States v. Jacobsen, 466 U.S. 109 (1984); United States v. Place, 462 U.S. 696 (1983). 111. See Jacobsen, 466 U.S. at 122-24; Place, 462 U.S. at 706-07. 112. See Amar, Fourth Amendment, supra note 2, at 768-69, 783 & n.97. 113. See Jacobsen, 466 U.S. at 122-23 & nn.22-23 (testing to determine whether substance is cocaine "does not compromise any legitimate interest in privacy"). See generally Arnold H. Loewy, The FourthAmendment as a Device for Protectingthe Innocent, 81 MICH. L. REV. 1229 (1983) (generalizing the insight).

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

line of defense: judges who see that extended pretrial detention might cause irreversible loss of key exculpatory evidence must order the timely release of the detainee, or soften the conditions of confinement, to prevent the anticipated loss from ever occurring. What happens, however, if this first line of defense fails, and evidence has already been irreversibly lost, because of the defendant's constitutionally overlong detention? At this point, prevention now means preventing an unreliable trial from occurring-dismissing with prejudice if necessary. Such a dismissal fails to remedy a past bodily liberty or reputation violation, but it does prevent a future reliable trial violation. The wrong here is the trial itself, if it poses an undue threat of an erroneous conviction owing to the government's past lapses.11 4 Dismissal with prejudice is indeed an exclusionary rule of sorts, but one designed to protect innocence. And that kind of exclusionary rule draws strong structural support from other constitutional provisions. As we have seen, the Fourth Amendment, rightly read, contains no exclusionary rule; but the Fifth contains several, all of which are designed to protect innocence and/or to prevent vexation and oppression at trial itself. The SelfIncrimination Clause excludes items from a criminal trial-a defendant's compelled statements-but as Ren6e Lettow and I have explained elsewhere, this exclusionary rule is designed to reduce the risk of erroneous convictions of innocent defendants." 15 Consider next the Double Jeopardy Clause, which excludes a second trial-dismisses with prejudice in effect-if a defendant has already been acquitted or convicted of the same offence. If the defendant has already been acquitted, a second trial would indeed pose an intolerable risk to innocence. If a second trial is okay, why not a third? If a third, why not a fourth, and so on? Eventually, the government may be able to wear an innocent defendant down, and find one statistically aberrant jury that would erroneously convict. Under this heads-we-win-tails-let's-do-it-over 1 6 regime, the obvious innocence-protecting spirit of Winship' would be undermined. If, by contrast, a defendant has already been convicted of the very same offense, what purpose could a second trial have, other than to

114. If, pretrial, it is unclear whether a trial itself would be unduly unreliable because of pretrial detention, a judge could hold the trial. At trial, the amount of unreliability caused by delay should be much easier to measure than it was to estimate pretrial; and the judge could grant a proper motion to dismiss during or at the end of the trial itself. See United States v. MacDonald, 435 U.S. 850, 858-59 (1978) (recommending this approach). 115. See Amar & Lettow, Fifth Amendment, supra note 3, at 900-01. In a nutshell, we argue that the best theory underlying the Self-Incrimination Clause is that many defendants, if forced to take the stand, might look bad and hurt their own cases even though they are innocent. 116. In re Winship, 397 U.S. 358, 362-64 (1970) (requiring proof beyond reasonable doubt in criminal cases).

1996]

SIXTH AMENDMENT FIRST PRINCIPLES "

vex him?117 Alongside the Double Jeopardy Clause's textual protections against multiple prosecutions for the "same offence," the due process principle prohibits innocence-threatening and vexatious multiple prosecutions more generally. The due process principle of collateral estoppel is a cousin of the double jeopardy idea of autrefois acquit:1 8 if a defendant prevails on

any factual issue in one criminal case, the government may not try to force him to prove it all over again in a second criminal case. Thus, although kidnapping and bank robbery are plainly different offenses, if a defendant is acquitted of kidnapping on the theory that the police simply nabbed the wrong man in a case of mistaken identity, the government would be barred from later trying him on a bank robbery charge growing out of the same episode. So too, a due process cousin of the double jeopardy idea of

117. In one context, a defendant can be tried after conviction: If his conviction is overturned because of government-induced error, he generally may be tried again. It's tempting to claim that retrial is allowed because a defendant waives his double jeopardy claim when he takes an appeal, but this will not wash. The system forces the defendant to "waive" this objection as a condition of allowing his appeal in the first place-an appeal that seeks to undo a government-induced error. The true logic allowing retrial is that a government acting in good faith should be allowed one fair trial, with a chance to prove guilt and win a conviction that will stick. See Peter Westen & Richard Drubel, Toward a General Theory of Double Jeopardy, 1978 SuP. CT. REV. 81, 102-06, 125-28. To vindicate that interest we force some defendants to run the traumatic trial gauntlet twice, through no fault or real choice of their own. A fortiori, the government should be allowed to try Bill once, even if it has already erroneously subjected him to an overlong and traumatic accusation period. In asserting the contrary, the Court's wooden and conclusory opinion in Strunk failed to ponder the implications of double jeopardy doctrine for highly analogous speedy trial issues. See Strunk v. United States, 412 U.S. 434, 438-39 (1973). Doctrinally, the Strunk Court could have applied the lesson of double jeopardy law not via "waiver" doctrine, but by holding that some delays between indictment and trial are, for fair trial purposes, "harmless error." See Chapman v. California, 386 U.S. 18, 22 (1967) (discussing various constitutional errors that may not require automatic reversal); Dickey v. Florida, 398 U.S. 30, 55 (1970) (Brennan, J., concurring) (noting possible relevance of Chapman to speedy trial issue). For an approach directly at odds with Strunk, see United States v. Ewell, 383 U.S. 116, 121 (1966): [This Double Jeopardy rule] has been thought wise because it protects the societal interest in trying people accused of crime, rather than granting them immunization because of legal error at a previous trial, and because it enhances the probability that appellate courts will be vigilant to strike down previous convictions that are tainted with reversible error .... These policies, so carefully preserved in this Court's interpretation of the Double Jeopardy Clause [should not be] seriously undercut by the interpretation given the Speedy Trial Clause .... This Ewell passage was quoted in its entirety and heavily relied on in the post-Strunk case of United States v. Loud Hawk, 474 U.S. 302, 313 (1986). Stunk, by contrast, wrongly "grant[ed] ... immunization because of legal error at a previous [stage]." Ewell, 383 U.S. at 121; cf. Pollard v. United States, 352 U.S. 354, 362 (1957) ("Error in the course of a prosecution resulting in conviction calls for the correction of the error, not the release of the accused."). 118. Although the Court has at times denied that collateral estoppel is rooted in due process rather than the Double Jeopardy Clause, the logic of the leading collateral estoppel case, Ashe v. Swenson, 397 U.S. 436 (1970), belies this denial. See Amar & Marcus, supra note. 18, at 30-31.

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

autrefois convict will bar vexatious multiple prosecutions for different offenses, where the government can point to no legitimate reason for forcing a defendant to undergo two traumatic trials rather than one single consolidated trial.1 19 The underlying principle of all these constitutional exclusionary rules outside the Sixth Amendment is to protect innocence and/or prevent a trial that would itself be vexatious and oppressive. Structural analysis suggests that the same should be true for the Sixth Amendment itself. d. Dismissal With Prejudice as "The Only Possible Remedy?" How, in the end, are we to explain the modern Court's ahistorical embrace of dismissal with prejudice as the "only possible remedy?""12 The perverse gravitational pull of the so-called Fourth Amendment exclusionary rule is no doubt part of the story. However, as the Court itself has noted, the exclusionary rule ordinarily blocks certain items of evidence, but does not 12 prevent the government from using other evidence to prove guilt. ' Dismissal with prejudice, by contrast, altogether bars prosecution of the guilty. And in the one corner of exclusionary rule precedent where exclusion would indeed mean dismissal altogether-Frisbie and its progeny-the 122 Court has emphatically resisted exclusion's gravitational pull. More generally, our modern system of constitutional criminal procedure seems to welcome broad, upside-down exclusions of reliable evidence1 23 as "fruits" of under the Kastigargloss on the Self-Incrimination Clause, ' ' 22 and under the Mascoerced confessions and other "poisonous trees, siah doctrine," for example. Leading scholars of criminal procedure have applauded this system; and criminal defense attorneys have grown up under it. But virtually all of these upside-down exclusions rest on misreadings of constitutional text, history, and structure. Criminal defense attorneys may treat an upside-down world as inevitable-the only possible approach-but of course this is a terribly convenient view for them. They

119. For more analysis, see Amar & Marcus, supra note 18, at 30-38. 120. Strunk, 412 U.S. at 440; Barker v. Wingo, 407 U.S. 514, 522 (1972). 121. Barker, 407 U.S. at 522 (quoted supra text accompanying note 36); United States v. Blue, 384 U.S. 251, 255 (1966) (quoted supra note 97). 122. See Frisbie v. Collins, 342 U.S. 519, 522 (1952); see also supra note 97 and accompanying text. 123. See Kastigar v. United States, 406 U.S. 441, 453-62 (1972) (excluding fruits of immunized testimony). 124. See generally Yale Kamisar, On the "Fruits" of Miranda Violations, Coerced Confes-

sions, and Compelled Testimony, 93 MIcH. L. REV. 929 (1995) (championing these exclusionary rules). 125. Massiah v. United States, 377 U.S. 201, 207 (1964) (excluding reliable recitation of defendant's uncoerced statements). The Massiah Court thought the Sixth Amendment right of counsel was somehow implicated by the facts at hand, but it is hard to see how. See id. at 209 (White, J., dissenting) ("Massiah was not prevented from consulting with counsel as often as he wished. No meetings with counsel were disturbed or spied upon.").

1996]

SIXTH AMENDMENT FIRST PRINCIPLES

are paid to get their clients off, their clients are often guilty, and upsidedown exclusion can help even guilty clients win-indeed, can help especially guilty clients win. Precisely because this at first seems so perverse, as nonlawyers intuit, it's convenient and comforting for lawyers to tell themselves that the Constitution compels this, and that there is no other way. Scholars should know better, but too few of those who write in Criminal Procedure do serious, sustained scholarship in Constitutional Law generally, or in fields like Federal Jurisdiction and Remedies. 126 As a result, discourse in constitutional criminal procedure has evolved separately, cutting itself off from larger themes of constitutional, remedial, and jurisdictional theory. Standard constitutional law modalities of text, history, and structure are often slighted, 127 and remedial lessons elsewhere in constitutional law ignored. Elsewhere in constitutional law, dismissal and exclusion are not the "only possible remedies." When freedom of speech or of the press is at stake, or equal protection, or due process, a Section 1983/Ex parte Young/Bivens128 model-featuring before-the-fact prevention via injunctions and after-the-fact compensation and deterrence via damagesreigns as the dominant remedial approach. This is the general remedial model I have tried to adapt to the knotty speedy trial issues at hand. After-the-fact damages are at times attacked as simply allowing the government to buy off constitutional rights with money, to cynically treat violations of sacred constitutional rights merely as the cost of doing business, to wrongly transform all constitutional rights into Takings-Clauselike "liability" rights. 129 This criticism is simply mistaken and rests on a misunderstanding of basic remedial theory. The first line of defense must always be prevention-here, by scheduling speedy trials; by issuing writs of habeas corpus to prevent undue detention; by quashing lingering indictments that stigmatize; by preventing irreparably unfair trials from ever occurring; and so on. And if an in-court trial 126. There are, of course, important counterexamples-such as the work of Tony Amsterdam, John Jeffries, Dan Meltzer, and Mike Seidman. See, e.g., Amsterdam, supra note 38; John C. Jeffries, Jr., Damagesfor Constitutional Violations: The Relation of Risk to Injury in Constitutional Torts, 75 VA. L. REV. 1461 (1989); Daniel J. Meltzer, Deterring Constitutional Violations by Law Enforcement Officials: Plaintiffs and Defendants as PrivateAttorneys General,

88 COLUM. L. REV. 247 (1988); Silas J. Wasserstrom & Louis M. Seidman, The Fourth Amendment as ConstitutionalTheory, 77 GEO. L.J. 19 (1988). 127. See generally PHILIP BOBBITr, CONSTITUTIONAL FATE (1982); Richard H. Fallon, Jr., A Constructivist Coherence Theory of ConstitutionalInterpretation, 100 HARV. L. REV. 1189

(1987). 128. See Bivens v. Six Unknown Named Agents of the Fed. Bureau of Narcotics, 403 U.S. 388 (1971); Exparte Young, 209 U.S. 123 (1908).

129. See U.S. CONST. amend. V ("[N]or shall private property be taken for public use, without just compensation .... "). The phrase "liability rights," of course, comes from the classic analysis of Guido Calabresi & A. Douglas Melamed, Property Rules, Liability Rules, and Inalienability: One View of the Cathedral, 85 HARV. L. REV. 1089 (1972). For important refinements, see Jules L. Coleman & Jody Kraus, Rethinking the Theory of Legal Rights, 95 YALE L.J. 1335 (1986).

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

right has been violated in a way that might have led to an unjust and erroneous conviction of guilt, 130 and in a way that a new trial could avoid, a sound remedial model would of course not say: "Too bad, Mr. Gideon, you were denied your express right to counsel, and you were thereby convicted

of a felony, quite possibly erroneously. Here's a few bucks, good luck in prison." Rather, the state must try to run the trial over again, this time without the constitutional violation. This re-running of trials is merely an adaptation of the first line of defense-prevention-and a reflection of the value our Constitution rightly places on innocence protection. But if a constitutional violation has already occurred out of court, a court

cannot really prevent it: a court can re-run a trial but it cannot turn back the clock of time and re-run the world outside the court. A court cannot give Andy back the five months he was unlawfully held pretrial, or make the five unconstitutional months of mud on Bill's name vanish nunc pro tunc. After-the-fact damages cannot metaphysically undo these harms, but neither can dismissal with prejudice.13 ' Nothing can, metaphysically. But our general law of remedies does countenance after-the-fact damages to compensate, as best we can, for the past injury done, and to deter-to prevent-

future injury. Indeed, the permissibility of punitive damages shows how the Takings

Clause objection is precisely inapt. When government takes a piece of property and pays a fair price for it, no punitive damages are ever awarded. Indeed, no right has ever been violated-the right is simply to after-thefact compensation.' 3 2 And so a judge should never enjoin a taking when the government stands ready to pay, cash in hand. By contrast, here we are focusing on rights that have been violated, where before-the-fact injunc130. This qualification, I submit, is the root idea underlying various cases and doctrines. See, e.g., Lockhart v. Fretwell, 506 U.S. 364, 372 (1993) (unavailability of ineffective assistance of counsel claim when lawyer's mistake does not result in fundamentally unfair or unreliable trial); McCleskey v. Zant, 499 U.S. 467, 494 (1991) (actual innocence and miscarriage of justice exceptions to abuse of the writ doctrine in habeas); Teague v. Lane, 489 U.S. 288, 313 (1989) (exception to rule of nonretroactivity in habeas for new rules "without which the likelihood of an accurate conviction is seriously diminished"); Murray v. Carrier, 477 U.S. 478, 495-96 (1986) (actual innocence and miscarriage of justice exceptions to procedural default in habeas); United States v. Bagley, 473 U.S. 667, 678 (1985) (material omission standard for prosecutorial nondisclosure); Stone v. Powell, 428 U.S. 465, 494 (1976) (unavailability of habeas to review exclusionary rule error); Chapman v. California, 386 U.S. 18, 22 (1967) (harmless error). 131. In a case where the only criminal penalty is a fine, isn't dismissal with prejudice similar to damages: an after-the-fact financial boon (here, via a fine not charged) for a pretrial loss of liberty or reputation? The difference, of course, is that an explicit damage scheme can be expressly tailored (via compensatory damages) to the precise legal injury caused in the past and (via punitive damages) to the precise need for future deterrence. Like the exclusionary rule, dismissal almost never achieves the right measure of compensation and deterrence, and if it does so, it is only by the wildest of coincidences-like a broken clock that gets the time right twice a day. 132. See Coleman & Kraus, supra note 129 (analyzing legal rights underlying liability rules).

1996]

SIXTH AMENDMENT FIRST PRINCIPLES

677

tions should have issued. Punitive damages are at times appropriate precisely to deter-to prevent-future violations. Indeed, the very concept of punitive damages originally entered into Anglo-American law in landmark proto-Fourth Amendment cases.1 33 The Framers' remedial system for constitutional wrongs was based solidly on punitive damages, not exclusions as the "only possible remedy." Until speedy trial discourse takes account of all this, glib judicial pronouncements about dismissal should not command the respect of thoughtful students of the Constitution. III. IN PARTICULAR: PUBLIC TRIAL

The next cluster of clauses in the Sixth Amendment focuses on the public, the people, and the community: "[T]he accused shall enjoy the right to a speedy and public trial, by an impartial jury of the State and district wherein the crime shall have been committed, which district shall have been previously ascertained by law ..... Both syntactically and substantively, these words reach .back to the Speedy Trial Clause and also push beyond it. The result is a distinctly republican vision of innocence protection and truth-seeking. A. THE GALLERY: A PUBLIC TRIAL

Textually, the right to a "public trial" tightly intermeshes with the right to a "speedy ... trial" and the two rights share much in common. As we have seen, the speedy trial right was crafted with the innocent man as the paradigm-a man falsely "accused" of "infamous" and "criminal" misconduct who would naturally want a "speedy trial" to "answer" the indictment and make his "defence." For the same reasons an innocent man might naturally want a "speedy" trial, he might want a "speedy and public" trial: he has nothing to hide, and indeed, wants only to clear his name in open court, with the bracing sunshine of publicity helping to dry off the mud on his name. When he wins, as he deserves to, he wants the world-the public-to know, so that he can get back his good name in civil society. In a case of malicious prosecution-say, trumped up charges against a vocal government opponent-a public trial can expose corruption for all to see. In a case of mistaken identity, a public trial may reduce the risk of future mistakes. And if a defendant committed the acts he is charged with, but believes them justified, he should want the community to hear and understand his reasons. Guilty defendants as a whole, by contrast, may be less enthusiastic about public trials, just as they may be less enthusiastic about speedy ones. An unspeedy trial may help many a guilty defendant; and so too with a secret 133. See Amar, Fourth Amendment, supra note 2, at 814-15 & nn.214-15.

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

trial, where bribery may be easier. But, as we have seen, the Constitution does not in general confer on the defendant a right to an unspeedy trial; and the same holds true for an unpublic trial. Indeed, the words that the

Court has used in the speedy trial context apply a fortiori to the public trial right: "[T]here is a societal interest in providing a speedy trial which exists separate from, and at times in opposition to, the interests of the accused. ' , 134 Begin with the text. Though the Sixth Amendment says the "accused" shall enjoy a right to a "public trial," it does not say that the accused has a right to a "secret" or "private" trial. Historically, virtually all criminal trials in England and America have been trials open to the people. 35 In his celebrated Commentarieson the Constitution,Joseph Story noted that the Sixth Amendment "does but follow out the established course of the common law in all trials for crimes. The trial is always public."'136 And so, to return to the text of the Sixth Amendment, the right of public trial is indeed a right of the "accused" and only the "accused," in the sense that he may waive trial altogether by pleading guilty.13 7 But if he pleads not guilty, and thus demands a trial, he must get a public trial, whether he will or no; for a trial from which the people are excluded is, in the Anglo-American tradition, not a "trial" at all.' 38 134. Barker v. Wingo, 407 U.S. 514, 519 (1972). 135. In re Oliver, 333 U.S. 257, 266 (1948); Gannett Co. v. DePasquale, 443 U.S. 368, 414, 420-21 (1979) (Blackmun, J., concurring in part and dissenting in part). See also infra note 138 and sources cited therein. 136. 3 JOSEPH STORY, COMMENTARIES

ON THE CONSTITUTION OF THE UNITED STATES

§ 1785, at 662 (1833) (emphasis added). 137. See supra note 23. 138. In his influential Institutes, Sir Edward Coke declared that the very word "court" implied public access: "[AlIl Causes ought to be heard, ordered, and determined before the judges of the kings courts openly in the kings courts, wither all persons may resort; and in no chambers, or other private places: for the judges are not judges of chambers, but of courts, and therefore in open court." COKE, supra note 72, at 103. (emphasis added); see also Richmond Newspapers, Inc. v. Virginia, 448 U.S. 555, 566-67 (1980) (plurality opinion) (" '[O]ne of the most conspicuous features of English justice, that all judicial trials are held in open court, to which the public have free access, . . . appears to have been the rule in England from time immemorial.' ") (quoting EDWARD JENKS, THE BOOK OF ENGLISH LAW

73-74 (6th ed. 1967)); id. at 597 (Brennan, J., concurring in judgment) (noting that "public access is an indispensable element of the trial process itself"); id. at 599 (Stewart, J., concurring in judgment) ("With us, a trial is by very definition a proceeding open to the press and to the public."); Oliver, 333 U.S. at 267 n.14 (" 'By immemorial usage, wherever the common law prevails, all trials are in open court, to which spectators are admitted.' ") (quoting 2 JOEL P. BISHOP, NEW CRIMINAL PROCEDURE § 957 (2d ed. 1913)); Craig v.

Harney, 331 U.S. 367, 374 (1947) ("A trial is a public event."); Bridges v. California, 314 U.S. 252, 271 (1941) ("The very word 'trial' connotes decisions on the evidence and arguments properly advanced in open court."). To be sure, at some point the general innocence-protecting principles of the Due Process Clause and the Sixth Amendment will influence the precise mode in which a trial must be public. At some extreme point, a trial that is too speedy could become a farce, see supra text accompanying note 82, and so too, a trial that is too public could become a circus. See Moore v. Dempsey, 261 U.S. 86, 91 (1923) (holding that verdict "produced by mob demonstration" is contrary to due process of law); cf. Estes v. Texas, 381 U.S. 532, 545-50 (1965) (enumerat-

1996]

SIXTH AMENDMENT FIRST PRINCIPLES

Structural analysis helps to identify some of the special purposes served by "public" trials in America. The phrase "the people" appears in no less than five of the ten Amendments that make up our Bill of Rights; 139 and so we would do well to take seriously the republican and populist overtones of its etymological cousin, "public," in a sixth-the Sixth-Amendment. Ours is a system of republican governments, state and federal--of governments of, by, and for the people.1 4 ° Here, the people rule-not day to day, but ultimately, in the long run. All governmental policy and governmental policymakers can, in time, be lawfully replaced by the sovereign people via ordinary elections and constitutional conventions. This ultimate right of the public to change policy and policymakers creates a strong presumption that governmental action in all three branches will be open to public scrutiny. As Justice Blackmun put the point: Judges, prosecutors, and police officials often are elected or are subject to some control by elected officials, and a main source of information about how these officials perform is the open trial. And the manner in which criminal justice is administered in this country is in and of itself of interest to all citizens. In Cox Broadcasting Corp. v. Cohn, ... it was noted that information about the criminal justice system "appears to us to be of critical importance to our type of government in which 4 the citizenry is the final judge of the proper conduct of public business."' The people, however, do not need to wait until Election Day to make a difference; their very presence in the courtroom can help discourage judicial misbehavior. As Sir Matthew Hale wrote in his widely influential treatise, "if the judge be PARTIAL, his partiality and injustice will be evident to all by-standers. ' ' 142 Or as Sir William Blackstone wrote in his even more widely influential treatise: "[Objections to evidence] are publicly stated, and by the judge are openly and publicly allowed or disallowed, in the face of the country; which must curb any secret bias or partiality, that might arise in his own breast.",143 The ability of the public to ing situations in which televising trial might prejudice proceedings). For a particularly tart description of the circus problem, and the suggestion that the newspaper media can typically represent the public in the courtroom itself, see Max Radin, The Right to a Public Trial, 6 TEMP. L.Q. 381 (1932). 139. See U.S. CONST. amends. I, II, IV, IX, X. 140. See generally Akhil Reed Amar, The Central Meaning of Republican Government: Popular Sovereignty, Majority Rule, and the DenominatorProblem, 65 U. COLO. L. REV. 749 (1994). 141. Gannett Co. v. DePasquale, 443 U.S. 368, 428-29 (1979) (Blackmun, J., concurring in part and dissenting in part). 142. MATTHEW HALE, THE HISTORY OF THE COMMON LAW OF ENGLAND 344 (6th ed.

1820). 143. 3 BLACKSTONE, supra note 52, at *372. Though this passage occurs in the context of a discussion of evidence law in civil cases, Blackstone elsewhere makes clear that the same principles apply to criminal cases. See 4 id. at *350.

680

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

judge the judge should tend to protect innocent defendants from judicial corruption or oppression,'" but public scrutiny is bad news for many a guilty defendant, who might prefer an incompetent judge, or one "partial" to the defendant's cause-an old political friend, perhaps, or a new financial one. So too, the public right to monitor witnesses at trial was designed to help the truth come out, and truth of course helps innocent defendants more than guilty ones, as a rule. If, at trial, a bystander happens to have relevant information bearing on a key point, he can bring the matter to the attention of court and counsel.' 4 5 In part because of this, witnesses who testify are less likely to perjure themselves in front of a public gallery-or at least this was the theory underlying the common law's commitment to public trials. In 1685, Solicitor General John Hawles put the point as follows: [T]he reason that all trials are public, is, that any person may inform in point of fact, though not subpoena'd, that truth may be discovered in civil as well as criminal cases. There is an invitation to all persons, who can inform the court concerning the146matter to be tried, to come into the court, and they shall be heard.

Truth was also Blackstone's theme: This open examination of witnesses viva voce, in the presence of all mankind, is much more conducive to the clearing up of truth, than [a] private and secret examination .... [A] witness may frequently depose which he will be ashamed to testify in a public and solemn that in private, 7 tribunal.

14

In short, the public trial was designed to infuse public knowledge into the trial itself, and, in turn, to satisfy the public that truth had prevailed at trial. A public trial would protect innocence, but would make life more difficult for the guilty. All these values have been turned upside down by modern doctrines that-in the name of the Constitution, no less!-exclude

144. See In re Oliver, 333 U.S. 257, 270 (1948) ("The knowledge that every criminal trial is subject to contemporaneous review in the forum of public opinion is an effective restraint on possible abuse of judicial power."). 145. See 6 JOHN H. WIGMORE, EVIDENCE § 1834 (J. Chadbourn rev. ed. 1976); Gannet Co., 443 U.S. at 383; id. at 427 (Blackmun, J., concurring in part and dissenting in part); Oliver, 333 U.S. at 270 n.24. 146. Sir John Hawles, Remarks upon Mr. Cornish's Trial, in 11 HOWELL'S STATE TRIALS 455, 460 (London, Hansand 1811).

147. 3 BLACKSTONE, supra note 52, at *373. For very similar language, see note 142, at 345.

HALE,

supra

19961

SIXTH AMENDMENT FIRST PRINCIPLES

evidence the public knows to be true. Put differently, even if a judge can prevent the jury from learning of some fact, she often cannot prevent the larger public from, at some point, learning of it. And the gap between public truth and truth allowed in the courtroom can demoralize the public, whose faith in the judicial system is a key goal of the public trial ideal.14 8 In the wise words of Professor Nesson: [A] verdict of not guilty or not liable will only undermine the legal system's projection of behavioral norms if the public has an independent basis for believing that the defendant did in fact commit the wrongful act.... Supporters of the [exclusionary] rule who argue that the rule sets few criminals free ... fail to appreciate the demoralizing message conveyed to the public when the assertion of an evidentiary rule that impedes the search for truth is permitted to override the substantive 14 9 norm embodied in criminal law. Thus, various modern exclusionary rules are not merely indefensible as a matter of text, history, and structure, and remedially inapt to boot. These modern upside-down rules also do violence to the elaborate adjudicatory architecture of the truth-seeking, confidence-enhancing, innocenceprotecting, public trial envisioned by the Sixth Amendment. B. THE JURY: TRIAL BY THE PEOPLE

Closely linked to the public trial idea is the jury trial idea. Here too, the relevant legal interests are not merely the accused's but also the people's. Here too, public participation in the criminal justice system was designed to enhance public legitimacy of the criminal justice system, to pursue truth, and to protect innocence. No idea was more central to our Bill of Rights than the idea of the jury. The only right secured in all state constitutions penned between 1776 and 1787 was the right of jury trial in criminal cases; 5 ' and even though the original Constitution omitted a general Bill of Rights, it did expressly protect criminal juries in Article 111.151 In the Bill of Rights itself, three

148. See Richmond Newspapers, Inc. v. Virginia, 448 U.S. 555, 571-72 (1980) (plurality opinion) (discussing how public trial can promote "public acceptance of both the [judicial] process and its results" and "confidence in the fair administration of justice") (quoting State v. Schmit, 139 N.W.2d 800, 807 (Minn. 1966)); id. at 595 (Brennan, J., concurring in the judgment) ("Public access is essential, therefore, if trial adjudication is to achieve the objective of maintaining public confidence in the administration of justice."); Gannett Co., 443 U.S. at 429, 448 (Blackmun, J., concurring in part and dissenting in part) (similar). 149. Charles Nesson, The Evidence or the Event? On JudicialProofand the Acceptability of Verdicts, 98 HARV. L. REV. 1357, 1367 n.31 (1985). 150. See Amar, supra note 23, at 1183. 151. See U.S. CONST. art. III, § 2, cl. 3 ("The Trial of all Crimes, except in Cases of Impeachment, shall be by Jury .... ").

682

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

separate Amendments explicitly safeguarded juries-grand, petit, and civilIE 2-and several other Amendments tightly intermeshed with the jury

idea. 153 At root, jury trials were, in Thomas Jefferson's words, "trials by the people themselves.' ' 1 54 And this right of trial by the people was not merely a right of the accused, but a right of the people-of the jurors too. To be sure, the Sixth Amendment says that the accused shall enjoy the right of "public trial, by an impartial jury." But, once again, the accused is nowhere given a right to a nonjury trial. The undiminished language of Article III is clear and emphatic: "The Trial of all Crimes ... shall be by Jury.'" s And so, just as a secret trial is no trial at all, a judge sitting without a criminal jury-at least at the federal level-is no court, and thus cannot try anything. The accused, and only the accused, can decide to plead guilty and thus waive trial altogether; but if he insists on standing trial, both judge and jury must be present. This, at least, is the theory of our Constitution, and of the Supreme Court throughout the nineteenth centuries,' 5 6 though it has been rejected by the modand early twentieth 57 ern Court.' An analogy to federal legislative bicameralism helps illustrate the point: a federal judge sitting without a jury is simply not a federal court capable of trying criminal cases, just as the Senate sitting without the House is not a federal legislature capable of passing laws. 1 58 Though this bicameral analogy rings odd in modern ears, it is precisely the one many supporters of a Bill of Rights had in mind in debates over the ratification of the federal Constitution. Here are the words of a leading Anti-Federalist

152. See U.S. CONST. amend. V (grand jury); id. amend. VI (petit jury); id. amend. VII (civil jury). 153. For much more discussion, see Akhil Reed Amar, Reinventing Juries: Ten Suggested Reforms, 28 U.C. DAvis L. REV. 1169 (1995); Amar, supra note 23, at 1182-99. In light of my extensive discussion of juries in these essays, I shall devote less space here to the jury than its intrinsic importance would otherwise dictate. 154. Letter from Thomas Jefferson to David Humphreys (Mar. 18, 1789), reprinted in THE PAPERS OF THOMAS JEFFERSON, 1788-89, at 676, 678 (Julian P. Boyd ed., 1958) [hereinafter THE PAPERS OF THOMAS JEFFERSON].

155. U.S. CONST. art. III, § 2, cl. 3 (emphasis added). 156. See Insurance Co. v. Morse, 87 U.S. (20 Wall.) 445, 451 (1874) (stating that criminal defendant must be tried by jury); Thompson v. Utah, 170 U.S. 343, 353-54 (1898) (holding that criminal defendant must be tried by court and twelve-person jury); cf. Schick v. United States, 195 U.S. 65, 67 (1904) (allowing defendant to waive right to jury in case involving petty offense, which Court found was not covered by Article III mandate); see also Callan v. Wilson, 127 U.S. 540, 549 (1888) (noting that Sixth Amendment was not "intended to supplant" Article III mandate). 157. See Patton v. United States, 281 U.S. 276, 299 (1930), criticized in Amar, supra note 23, at 1196-99. 158. Of course, just as the Senate may act without the House in some areas-treaties, confirmations, expelling its own members, and so on-so a judge may act without a jury in some areas, such as accepting guilty pleas, setting bail, and sentencing. I

1996]

SIxTH AMENDMENT FIRST PRINCIPLES

pamphleteer: It is essential in every free country, that common people should have a part and share of influence, in the judicial as well as in the legislative department.... The trial by jury in the judicial department, and the collection of the people by their representatives in the legislature ... have procured for them, in this country, their true proportion of influence .... 159 So too, another leading Anti-Federalist defined the jury as "the democratic branch of the judiciarypower-more necessary than representatives in the legislature."1 6 Thomas Jefferson emphatically agreed: "[lit is necessary to introduce the people into every department of government .... Were I called upon to decide whether the people had best be omitted in the Legislative or Judicial 1department, I would say it is better to leave them 61 out of the Legislative.' Just as the House could monitor and expose--check-any partiality or corruption in the Senate, so the jurors on the lower bench could check a corrupt or partial set of judges on the upper bench. According to Jefferson: [W]e all know that permanent judges acquire an Esprit de corps, that being known they are liable to be tempted by bribery, that they are misled by favor, by relationship, by a spirit of party, by a devotion to the Executive or Legislative.... It is left therefore to the juries, if they think the permanent judges are under any biass whatever in any cause, to take upon themselves to judge the law as well as the fact. They never exercise this power but when they suspect partiality in the judges .... 162 Even Alexander Hamilton, who disagreed with Jefferson about a great deal, and pooh-poohed the need for an explicit Bill of Rights, 163 agreed that juries discouraged corruption. In discussing the civil jury in The FederalistNo. 83, Hamilton wrote: The strongest argument in its favor is that it is a security against corruption. As there is always more time and better opportunity to tamper with a standing body of magistrates than with a jury summoned for the occasion, there is room to suppose that a corrupt influence would more

159. Letters From the Federal Farmer (IV), in 2 THE COMPLETE

ANTI-FEDERALIST

249-50

(Herbert J. Storing ed., 1981) [hereinafter THE COMPLETE ANTI-FEDERALIST]. 160. Essays by a Farmer (IV), in 5 id. at 38. 161. Letter from Thomas Jefferson to L'Abbe Arnoux (July 19, 1789), reprintedin 15 THE PAPERS OF THOMAS JEFFERSON, supra note 154, at 282-83. 162. Id. 163. See generally THE FEDERALIST No. 84 (Alexander Hamilton).

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

easily find its way to the former than the latter.... The temptations to prostitution which the judges might have to surmount must certainly be much fewer, while the co-operation of a jury is necessary, than they might be if they had themselves the exclusive determination of all causes.164 In short, like the people in the gallery box, the people in the jury box at a public trial could detect and deter judicial misconduct. A guilty defendant looking to bribe his way out, or pull a few strings, might well prefer a closed bench trial, and so might a judge and prosecutor on the take. But the Constitution did not permit the defendant, even with the agreement of judge and prosecutor, to oust the people from their rightful places in both the gallery box and the jury box. For the Framers, however, the criminal jury was much more than an incorruptible factfinder. It was also, and more fundamentally, a political institution embodying popular sovereignty and republican self-government. Through jury service, citizens would learn their rights and duties, and actively participate in the governance of society. In the words of the prominent Anti-Federalist essayist, "Federal Farmer": "[The people's] situation, as jurors and representatives, enables them to acquire information and knowledge in the affairs and government of the society; and to come forward, in turn, as the centinels and guardians of each other."16' 5 Jury service was both a duty and a right-a badge of first class citizenship, no less than the right to vote. Indeed, throughout American history and constitutional discourse, the right to vote and the right to serve on juries have stood as siamese twins, joined at the hip. After all, in deciding guilt or innocence, jurors vote-that is what they do-and historically, ordinary voters have been eligible to serve on juries.' 6 6 Tocqueville put it well: The jury system as understood in America seems to me as direct and extreme a consequence of the ... sovereignty of the people as universal suffrage. They are both equally powerful means of making the majority prevail ....[Tihe jury is above all a political institution [and] should be

made to harmonize with the other laws establishing the sovereignty.... [Flor society to be governed in a settled and uniform manner, it is essential that the jury lists should expand or shrink with the lists of [iun America all citizens who are electors have the voters.... [In general] 67 right to be jurors. 1 164.

THE FEDERALIST

No. 83, at 500-01 (Alexander Hamilton) (Clinton Rossiter ed.,

1961). 165. Letters From the Federal Farmer (IV), in 2 THE COMPLETE ANTI-FEDERALIST, supra note 159, at 250. 166. See generally Vikram D. Amar, Jury Service as PoliticalParticipationAkin to Voting, 80 CORNELL L. REV. 203 (1995) [hereinafter Vikram D. Amar, Jury Service]. 167. ALEXIS DE TOCQUEVILLE, DEMOCRACY IN AMERICA 273, 728 (Jacob P. Mayer ed.,

1969).

1996]

SIxTH AMENDMENT FIRST PRINCIPLES

This populist vision of the people's right to vote and serve on juries may at times trump a given defendant's desires. Juries at the Founding were supposed to represent the polity-the people-not the defendant; and subsequent constitutional amendments have made clear that our polity now includes those once excluded-blacks and women, for example. Both the plain and the deep meaning of these amendments is that the right to vote-on juries too-may not be abridged on the basis of race or sex or class or age.' 68 And this right is paramount even if a defendant prefers otherwise, and wants a jury that looks like him rather than like America. Here the recent Supreme Court has been on just the right track, striking down race-based exclusions from jury service, even when a defendant seeks these exclusions via peremptory challenges. 16 9 The role of the criminal jury, however, involves even more than reliable factfinding and republican self-government. It also involves normative judgment. In this last respect, the criminal jury wields more power than its civil counterpart. Criminal trials are unavoidably morality plays, focusing on the defendant's moral blameworthiness or lack thereof. And the assessment of his moral culpability is, under the Sixth Amendment, a task for the community, via the jury, and not the judge-but with an innocenceprotecting twist. No judge can ever find a defendant guilty "as a matter of law," no matter how clear the defendant's factual and moral guilt to the judge.!7" No man who claims innocence can be condemned as guilty unless the community, via the jury, pronounces him worthy of moral condemnation. (And once acquitted by a jury, the defendant is, under the clear logic of both the Sixth Amendment and the Double Jeopardy Clause, forever quit of the charge.)1 7 1 But a judge who finds a defendant innocent "as a matter of law" may set him free, even if the jury disagrees. A careful comparison of the Sixth and Seventh Amendments' descrip168. See U.S. CONST. amend. XV (race); id. amend. XIX (sex); id. amend. XXIV (class); id. amend. XXVI (age). See generally Vikram D. Amar, Jury Service, supra note 166. 169. See Powers v. Ohio, 499 U.S. 400, 406-09 (1991) (striking down prosecutorial racebased peremptories in a criminal case); Edmonson v. Leesville Concrete Co., 500 U.S. 614, 616, 625-26 (1991) (striking down race-based peremptories in a civil case); Georgia v. McCollum, 505 U.S. 42, 59 (1992) (striking down defendant's race-based peremptories in a criminal case). Nor are such limits on peremptories any violation of an accused's right to an impartialjury. Surely the trial judge and appellate panel-the upper house in our bicameral judiciary-should also be impartial, yet defendants have never enjoyed a constitutional right of peremptory challenge against judges. And so the Supreme Court has repeatedly and correctly held that peremptory challenges are in no way required by the idea of impartiality. See, e.g., Stilson v. United States, 250 U.S. 583, 586 (1919); Swain v. Alabama, 380 U.S. 202, 219 (1965); Batson v. Kentucky, 476 U.S. 79, 91 (1986); McCollum, 505 U.S. at 57. 170. See Westen & Drubel, supra note 117, at 124-34. 171. See id. at 124-32, 133 & n.241; Peter Westen, The Three Faces of Double Jeopardy: Reflections on Government Appeals of Criminal Sentences, 78 MICH. L. REV. 1001, 1012-25, 1033-34 & n.99 (1980); see also Gregg v. Georgia, 428 U.S. 153, 199 n.50 (1976) (plurality opinion) (linking Sixth Amendment and Double Jeopardy Clauses). For further documentation, see Amar & Marcus, supra note 18, at 57-58 & n.279.

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

tions of the jury's role is illuminating. The Seventh explicitly highlights the role of the civil jury in finding "fact[s]"; 172 the Sixth does not. To be sure, at the Founding, the dominant view among well-trained lawyers was that a jury, when rendering a general verdict, could take upon itself the right to decide both law and fact.' 7 3 So said Chief Justice Jay for a unanimous Supreme Court in 1794.174 But, on the civil side, several doctrinal devices in 1791 enabled judges to avoid or overturn general verdicts, 175 and these devices have only grown in number and power over the last 200 yearsnonsuits, demurrers, summary judgments, directed verdicts, special verdicts, special interrogatories, new trials, J.N.O.V.s, and so on. But parallel devices did not exist in criminal cases in 1791, and few have emerged since: every criminal jury verdict is a general verdict in which the jury decides law and fact "complicatedly," to use an eighteenth-century phrase. And no criminal jury verdict of acquittal can ever be overturned. By contrast, note how the Seventh Amendment explicitly permits limited "re-examin[ation]" of civil jury verdicts.' 7 6 C. THE DISTRICT: TRIAL BY THE COMMUNITY

This role of moral judgment is subtly accented by the little-discussed District Clause of the Sixth Amendment. Under the Sixth, but not the Seventh, the jury must generally be drawn from "the ... district wherein the crime shall have been committed." At first blush, this seems like a simple venue provision, designed merely for ease of litigation: the case should be tried where the crime occurred because that's where the witnesses and physical evidence are.17 7 There is much to be said for this simple view, and it nicely meshes both with the general truth-seeking mission of the Sixth Amendment, and the specific logic of a gallery to monitor witnesses and bring new facts to light.' 7 ' But if mere litigation

172. U.S. CONST. amend. VII ("[A]nd no fact tried by a jury, shall be otherwise reexamined in any Court of the United States, than according to the rules of the common law."); see Ian Ayres, Pregnant With Embarrassments:An Incomplete Theory of the Seventh Amendment, 26 VAL. U. L. REV. 385, 401 (1991). 173. See Amar, supra note 23, at 1193. 174. See Georgia v. Brailsford, 3 U.S. (3 DalI.) 1, 4 (1794).

175. See Edith G. Henderson, The Background of the Seventh Amendment, 80 HARV. L. REV. 289, 299-320 (1966); Rende B. Lettow, New Trial for Verdict Against Law: Judge-Jury Relations in Early Nineteenth-CenturyAmerica, 71 NOTRE DAME L. REV. 505 (1996). 176. See supra note 172.

177. Technically, the District Clause prescribes where the jurors must come from, rather than where they must sit at trial. See infra text accompanying note 179. But Article III, Section 2-which, as we have seen, dovetails with the Sixth Amendment-does speak of

venue (the place of the trial) rather than vicinage (the place from whence the jurors come): "The trial of all Crimes ... shall be held in the State where the said Crimes shall have been committed .. " See generally Drew L. Kershen, Vicinage, 30 OKLA. L. REV. 1 (1977). 178. For historical support for this vision, see, e.g., Letters From the Federal Farmer (IV), in 2 THE COMPLETE ANTI-FEDERALIST, supra note 159, at 249 (linking vicinage to ease

19961

SIxTH AMENDMENT FIRST PRINCIPLES

687

convenience was the only issue, why didn't the Seventh Amendment likewise call for civil jury trial in "the State and district wherein the cause of action shall have arisen?" What's more, many defendants might find a jury from the crime district quite inconvenient and uncongenial-imagine, for example, a travelling salesman who lives in one state tried for a crime that took place in a different state and district far from his home and friends. Finally, we should note that, technically, the District Clause regulates the place from which jurors are chosen, rather than the place that they must 179 sit at trial. Underlying the District Clause is also, perhaps, the idea that a crimeunlike a civil wrong-constitutes a moral rupture, a distinct breach of the peace of the place where the crime occurred."' 0 A crime is committed not

merely against the victim, but against the community. And an apt judicial response to this crime, this moral rupture, requires not merely good

factfinding, but moral judgment-moral judgment by the community via the jury."'

of assembling oral evidence and "cross examining witnesses" thereby leading "to the proper discovery of truth"); 2 DEBATES ON THE ADOPTION OF THE FEDERAL CONSTITUTION 110 (Jonathan Elliot ed., 1888) [hereinafter ELLIOT'S DEBATES] (remarks of Mr. Holmes in Massachusetts ratifying convention) (stating that the "local situation" of jury from the place of the crime would better enable them to "judge of the credibility of the witnesses"); see also William W. Blume, The Place of Trial of Criminal Cases: ConstitutionalVicinage and Venue, 43 MICH. L. REV. 59, 64-65 (1944) (quoting May 16 and 17, 1769 Virginia Resolves linking venue and vicinage ideas to concerns about "speedy Justice," pretrial detention, and fair trials; transporting an American defendant overseas for trial in England would prolong and harshen his detention-keeping him in "Fetters amongst Strangers" in a "distant land [with] no Friend, no Relation [to] alleviate his Distresses" and "no Witness[es] ... to testify [to] his Innocence"). 179. See 3 ELLIOT'S DEBATES, supra note 178, at 547 (remarks of Edmund Pendleton at Virginia ratifying convention). 180. Historically, crime has been considered peculiarly "local" in nature; at the Founding, a court would not enforce the criminal laws of another sovereign even though it would enforce the other sovereign's civil laws. See Drew L. Kershen, Vicinage, 29 OKLA. L. REV. 801, 811 (1976). 181. See Kershen, supra note 177, at 79-94. During the late colonial period, Americans strongly objected to the so-called Murderers' Act, passed by Parliament after the Boston Massacre. The Act provided "that any government or customs officer indicted for murder [in America] could be tried in England, beyond the control of local juries." JOHN M. BLUM ET AL., THE NATIONAL EXPERIENCE 95 (3d ed. 1973). This circumvention of the judgment of the victimized community was attacked as a "Mock Trial" system in the Declaration of Independence. See THE DECLARATION OF INDEPENDENCE para. 17 (U.S. 1776). Although English and American juries might differ in their factual findings of whodunit, they were even more likely to disagree about normative issues of excuse, provocation, justification, and self-defenseissues at the heart of the Boston Massacre trials. Note that in condemning the Murderer's Act, Americans were siding against certain defendants. The Sixth Amendment, by contrast, speaks of the rights of "the accused." Thus, the defendant can arguably waive the vicinage rules of the Amendment. But of course the defendant generally lacks a constitutional right to demand a different vicinage, unless due process and jury impartiality so require. See Amar, supra note 23, at 1197. (Query whether the more absolute venue mandate of Article III, quoted supra note 177, is also waivable.)

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

But here too, how can the jury judge well on behalf of the community if, because of upside-down exclusion rules, it is denied reliable information that is known to the general community?18 2 IV. IN PARTICULAR: FAIR TRIAL Finally, let us turn to the Sixth Amendment's closing cluster of clauses, protecting nothing less than a defendant's right to put on his defense-to show he didn't do it. Here too, truth-seeking and innocence protection loom large. A. NATURE AND CAUSE OF ACCUSATION

"[T]he accused shall enjoy the right ... to be informed of the nature and cause of the accusation." The need for this fundamental right is obvious: if a defendant doesn't even know what he is charged with having done, how can he show he didn't do it? Only slightly less obvious is the way in which this fundamental right is especially valuable for the wholly innocent defendant. An "accusation" after all, typically focuses on allegations about the defendant's past conduct: what he did, when, where, why, how, with whom, to whom, and so on. If the government is in fact on the nose in its accusation, the accusation itself may tell the guilty defendant only what he already knows. But if the accusation is way off base-say, in a case of mistaken identity-it may tell the innocent defendant a great deal about where the government went wrong, and how he might go about showing this at trial or before. B. CONFRONTATION

"[Tihe accused shall enjoy the right ... to be confronted with the witnesses against him." This clause builds on several of the clauses we have already examined. Like many other intermeshing Sixth Amendment ideas, confrontation is designed to promote the truth. First, like the Public Trial Clause, the Confrontation Clause may discourage deliberate perjury by prosecution 182. To the extent that government illegality might be relevant to the defendant's culpabil-

ity or to the appropriateness of moral condemnation of his conduct, jurors could decide to take illegality into account in rendering their moral verdict. But this is a scheme of evidentiary inclusion, not exclusion-and one that (by hypothesis) is linked to the defendant's normative culpability. For other proposals to give criminal juries a greater role in monitoring illegal searches and seizures, see Ronald J. Bacigal, A Casefor Jury Determination of Search and Seizure Law, 15 U. RICH. L. REV. 791 (1981); George C. Thomas & Barry S. Pollack, Saving Rights from a Remedy: A Societal View of the Fourth Amendment, 73 B.U. L. REV. 147 (1993).

1996]

SixTH AMENDMENT FIRST PRINCIPLES

689

witnesses, who might be ashamed to tell their lies with the defendant in the room, and afraid that their lies will not stand up to open scrutiny. Second, by simply allowing a defendant to hear a witness's story, the clause may help an innocent defendant to figure out where the witness might be mistaken (perhaps in all good faith). In this respect, the Confrontation Clause echoes and amplifies the themes of its "nature and cause" neighbor: unless a defendant knows what the government is alleging, how can he show he didn't do it, or show where the government went wrong? Third, the clause enables the defendant not merely to hear the witness's story, but to directly question and cross-examine it-to show the jury and the public where the holes are-and to invite the witness herself to supplement, or clarify, or revise the story, so that the jury and the public may hear the whole truth. In this last respect, the Confrontation Clause links arms with its other neighbor, the Compulsory Process Clause, which also affirms a defendant's right to present truthful evidence before the jury and the public. In light of the obvious linkages between public trial and confrontation, we should not be surprised that Blackstone discussed the ideas together (in a chapter on trial by jury), or that he stressed truth-seeking: The oath administered to the witness is not only that what he deposes shall be true, but that he shall also depose the whole truth ....And all this evidence is to be given in open court, in the presence of the parties, their attorneys, the counsel, and all bystanders, and before the judge and jury .... This open examination of witnesses viva voce, in the presence of all mankind, is ...conducive to the clearing up of truth ....Besides, the occasional questions of the judge, the jury, and the counsel, propounded to the witnesses on a sudden, will sift out the truth much better than a formal set of interrogatories previously penned and settled: and the confronting of adverse witnesses is also another opportunity of obtaining a clear3 discovery, which can never be had upon any other method of trial. 1 Modern Supreme Court caselaw has exuberantly echoed Blackstone here, defining the "Confrontation Clause's very mission" as promoting "the 183. See 3 BLACKSTONE, supra note 52, at *372-73; see also 4 id. at *356. Blackstone borrowed heavily, it seems, from Hale: [O]ftentimes witnesses will deliver [in private] that, which they will be ashamed to testify publicly.... [M]any times the very MANNER of delivering testimony, will give a probable

indication, whether the witness speaks truly or falsely.... [Cross-examination] beats and boults out the truth much better, ... and [is] the best method of searching and sifting out the truth. HALE, supra note 142, at 345.

690

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

accuracy of the truth-determining process in criminal trials";' and labelling cross-examination the "greatest legal engine ever invented for the ' discovery of truth."185 Though the text and purposes of the Confrontation Clause seem clear enough, modern Supreme Court caselaw on the clause is surprisingly

muddled in logic and exposition. The results, on the whole, are sensible enough, but the Court has had difficulty believing that the clause could really mean what it says. If witness A takes the stand at trial and testifies about what her best friend B told her about the defendant, the Court has

worried that this hearsay might impede the defendant's right to directly confront his true accuser: the out-of-court declarant, friend B. 186 On the other hand (the Court has reasoned), it would be utterly impractical to try to exclude all hearsay from criminal trials; surely, it makes sense to allow in some forms of hearsay.' 8 7 And so the Court has balanced. But the words of the Confrontation Clause do not seem to balance; they seem to state a bright-line rule. Thus the Court has decided the words cannot possibly

mean what they say-they merely state a principled preference for live testimony. 88 This interpretive strategy runs rampant in modem constitutional criminal procedure. Though the Fourth Amendment's words do not explicitly require warrants and probable cause for every search and seizure, the 1 89 But since these Court has at times assumed they do so impliedly.

implied rules, if taken literally, make no sense in many cases, the Court has balanced, at times treating the Amendment as merely creating a preference

for warrants19°-just as the Sixth (on the Court's account) creates a preference for live testimony. The problem, of course, is that the word "preference" nowhere appears in the Fourth and Sixth Amendments, or in their accompanying history.' 9 1 Consider next the Fifth Amendment's Double Jeopardy Clause. Though its words bar retrial after acquittal or conviction on the "same offence" the 184. See Tennessee v. Street, 471 U.S. 409, 415 (1985) (quoting Dutton v. Evans, 400 U.S. 74, 89 (1970) (plurality opinion)); see also United States v. Inadi, 475 U.S. 387, 396 (1986) (similar); Idaho v. Wright, 497 U.S. 805, 825 (1990) (similar). 185. California v. Green, 399 U.S. 149, 158 (1970) (quoting 5 JOHN H. WIGMORE, EvIDENCE § 1367); see also Kentucky v. Stincer, 482 U.S. 730, 736 (1987) (similar); White v. Illinois, 502 U.S. 346, 356 (1992) (similar); cf. Pointer v. Texas, 380 U.S. 400, 404 (1965) ("[P]robably no one ... would deny the value of cross-examination in exposing falsehood and bringing out the truth in the trial in a criminal case."). 186. See, e.g., Ohio v. Roberts, 448 U.S. 56, 63 (1980). 187. See id. 188. See id.; Maryland v. Craig, 497 U.S. 836, 849 (1990) (quoting Roberts and italicizing and emphasizing "preference"). 189. See, e.g., Coolidge v. New Hampshire, 403 U.S. 443, 454-55 (1971); Johnson v. United States, 333 U.S. 10, 13-15 (1948). 190. For examples and discussion, see Amar, Fourth Amendment, supra note 2, at 769-70. 191. There are many other problems with a "preference" for warrants. See id. at 762-81.

1996]

SIxTH AMENDMENT FIRST PRINCIPLES

Court has decided that "same" really cannot mean "same"-i.e., identical. Rather, it must mean "greater and lesser-included"-the so-called Blockburger test. 192 But sometimes literal application of this Blockburger test would lead to absurd results. Imagine a case where the government convicts a defendant of attempted murder. After conviction, the victim dies from her injuries. Shouldn't the government now be allowed to prosecute for murder (with a set-off for any earlier punishment to avoid double counting)? The Court has made clear that reprosecution can occur on these facts' 9 3-here too, it has balanced. But if the Double Jeopardy Clause really does state a bright-line rule, by what right do judges balance it away? Finally, consider the Fifth Amendment Self-Incrimination Clause. Though the words of the clause speak only of "witness[ing]," the Court has insisted that the clause bars physical and other fruit of compelled out-ofcourt statements. 194 But sometimes it makes good sense to require records, of business activity and the like, and then later use those records to help detect crime. And so here too, in its required records cases, the Court has simply balanced. 195 Yet the text of the clause seems more rule-like than this. The solution to all these problems, I have suggested elsewhere, begins with taking the text seriously. The words of the Fourth Amendment, properly read, do not require or prefer warrants.' 9 6 The word "same" in the Double Jeopardy Clause means what it says.' 97 The Self-Incrimination Clause does not bar fruit, but only certain types of "witness[ing]."' 9 8 The rules in the Fourth and Fifth Amendments make sense as rules. Of course, at times we must go beyond the letter of these rules to protect their spirit, and thwart governmental shams and evasions of fundamental rights. But we cannot properly understand the true principles underlying these rules until we understand the rules proper, and take them seriously as rules.' 99 As it turns out, this approach, which helps sort out the Fourth and the Fifth Amendments, also works well in the Sixth, and indeed, neatly solves the Court's current Confrontation Clause conundrum. The place to begin is the text-in particular, the word "witness." Hasn't the Court wrongly conflated the word "witness" in the Confrontation 192. See Blockburger v. United States, 284 U.S. 299, 304 (1932). 193. See, e.g., Illinois v. Vitale, 447 U.S. 410, 420 n.8 (1980); Brown v. Ohio, 432 U.S. 161, 169 n.7 (1977); Jeffers v. United States, 432 U.S. 137, 151-52 (1977); Diaz v. United States, 223 U.S. 442, 448-49 (1912).

194. See, e.g., Kastigar v. United States, 406 U.S. 441 (1972). 195. See, e.g., Baltimore City Dep't of Social Servs. v. Bouknight, 493 U.S. 549 (1990); California v. Byers, 402 U.S. 424 (1971); Shapiro v. United States, 335 U.S. 1 (1948). 196. See generally Amar, Fourth Amendment, supra note 2. 197. See generally Amar & Marcus, supra note 18. 198. See generally Amar & Lettow, Fifth Amendment, supra note 3. 199. For a similar approach, see Coy v. Iowa, 487 U.S. 1012, 1020-21 (1988).

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

Clause with the somewhat different idea of "an out-of-court declarant"or, more elaborately still, "an out-of-court declarant whose utterance is intro2 duced for the truth of the matter asserted"-under the hearsay rule? " In ordinary language, when witness A takes the stand and testifies about what her best friend B told her out of court, A is the witness, not B. Imagine, for example, that B were later asked whether she had ever before been a witness in a criminal prosecution. Surely B could say no; indeed, she may 20 1 not even know that witness A paraphrased her words on the stand. Of course, sometimes words in a legal document mean something different from the same words in ordinary language. However, in a Constitution ratified by, subject to, and proclaimed in the name of, the people, it would 2 °2 At be unfortunate if words generally could not be taken at face value. any rate, surely a careful ordinary citizen reading the Confrontation Clause look to see how the word is used and pondering the word "witness" might 20 3 elsewhere in the Constitution itself. Consider, then, the Treason Clause of Article III, Section 3, one of a handful of clauses in the original document identified by the Federalist 24 Papers as in the nature of a traditional bill of rights: "No Person shall be convicted of Treason unless on the Testimony of two Witnesses to the 20 5 Imagine that a same overt Act, or on Confession in open Court." defendant is on trial for treason, with his life in the balance, and witness A testifies that she saw defendant's overt act, and that friend B told A that B also saw the same act. A is of course a witness; but should out-of-court friend B count as a witness too, within the meaning of the Treason Clause? I should hope not. Indeed, it is hard to imagine a more patent (and, if permitted, potent) evasion of the words and the spirit of the Treason Clause's requirement of two witnesses. And so, here at least, "witness" most clearly does not mean any out-of-court declarant. Consider next the Fifth Amendment Self-Incrimination Clause: "No person ... shall be compelled in any criminal case to be a witness against himself." The core rule here seems clear enough: a defendant cannot be forced to take the stand in his own case. Thus, the core meaning of "witness" resonates with ordinary language, applying to those who take the stand and testify in open court, but not to all out-of-court declarants. But the Self-Incrimination Clause shows that we need to refine our 200. For a clear example of this error, see Ohio v. Roberts, 448 U.S. 56, 63 (1980). For a discussion of the "truth of the matter asserted" wrinkle, see Tennessee v. Street, 471 U.S. 409 (1985). 201. See infra note 212. 202. See 1 STORY, supra note 136, § 451, at 436-37. 203. For an exemplary illustration of this interpretive technique, see McCulloch v. Maryland, 17 U.S. (4 Wheat.) 316, 413-15 (1819). 204. See THE FEDERALIST No. 84, at 510-11 (Alexander Hamilton) (Clinton Rossiter ed., 1961). 205. U.S. CONST. art. III, § 3, cl. 1.

1996]

SIXTH AMENDMENT FIRST PRINCIPLES

definition of "witness" to vindicate the obvious intent of the rule. Consider the following question: is a person who "testifies" by videotaped deposition, or who prepares a written deposition or affidavit intended for court use and then used in court, a witness? In ordinary language, this question might be a close one, but a quick look at the Self-Incrimination Clause tells us that the obvious answer is yes. The government may not force the criminal defendant to take the stand and answer questions-that is the plain meaning of the clause. But suppose, in the middle of the trial, the government temporarily "adjourns" the proceedings. The prosecutor, with the help of the bailiff, forcibly takes the defendant into the next room and forces him, upon pain of contempt, to answer questions. The next morning, the prosecutor introduces a videotape of this interrogation, or a written transcript of it, or an affidavit or a deposition signed (again, under penalty of contempt) by the defendant. Surely this is an obvious violation of the core rule against Self-Incrimination, at least in spirit. But also, I believe, in letter; we should treat an affidavit or transcript, prepared for in-court use and introduced in court as testimony, as witnessing. And of course the same thing is true if the compelled interrogation yielding the deposition occurred the day before the trial, or a year before, rather than during the trial itself. Now turn to the Constitution's next use of the word "witness," in the Confrontation Clause itself. The core meaning, here too, is clear enough: when a witness in court takes the stand, the defendant must have a chance to look him in the eye and confront him with questions. And so, if the defendant were banished from the courtroom when witness A took the stand, the violation would be flagrant. But suppose a hamfisted government tried to move the mountain rather than Mohammed: in the middle of the trial, proceedings are "adjourned," and the prosecutor, jury, and judge all troop into the next room to hear witness A's story, while defendant Mohammed is obliged to stay put. This too, is an obvious violation. But now suppose instead that a more clever government adjourns the trial, walks into the next room and gets witness A to tell his story, and A immediately leaves the jurisdiction. When the trial resumes the next day, the prosecutor introduces a video transcript of witness A's story, or a written affidavit or deposition, as testimony. Surely this sneakiness violates the core rule of the Confrontation Clause, at least in spirit. But also, I submit, in letter. As in the Self-Incrimination Clause, we must properly read the word "witness" to encompass videotapes, transcripts, depositions, and affidavits when prepared for court use and introduced as testimony. And of course this is also true if the deposition was taken before the trial, rather than during it.2" 6 206. See Gannett Co. v. DePasquale, 443 U.S. 368, 437 (1979) (Blackmun, J., concurring in part and dissenting in part).

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

The case of depositions differs in key ways from the case where witness A takes the stand and recounts what her best friend B said. Our deposition was given in a formal, solemn setting, and typically under oath or affirmation; and so a jury might give it great weight on that account, treating it as equal to sworn testimony in the courtroom itself. (This was especially true in the Framers' world, where great weight was placed on oaths.)2 7 By contrast, a jury would be much more likely to discount friend B's tale, since B took no oath, and may have been speaking loosely, without knowledge of the grave legal stakes at issue. Second, a deposition purports to be a precise rendition of the deponent's testimony, once again encouraging a jury to treat it as equivalent to in-court testimony. By contrast, a jury would be less likely to view A's account as a precise repetition of B's words. Third, in depositions, the government has manipulated the process to get witness testimony, qua testimony, with all the formal trappings, while excluding the defendant. Government administers the oath, asks the questions, and transcribes the answers, while purposefully excluding the accused. No similar manipulation occurs when friend B talks to her best friend A, perhaps even before the crime has occurred, or the government has appeared on the scene. In light of all this, it makes sense to say that a deponent is a "witness" in a way that friend B is not. Our reading of the word "witness" fits its ordinary, everyday meaning, and closely follows the logic of the core rules of the Self-Incrimination and Confrontation Clauses. It also has several other virtues. For starters, it perfectly fits the history behind the Confrontation Clause, a history born of revulsion against trial by affidavit. 20 8 Note how Blackstone presented the confrontation right as designed to avoid the unfairness of governmentprepared depositions: This open examination of witnesses viva voce, in the presence of all mankind, is much more conducive to the clearing up of truth, than the private and secret examination taken down in writing before an officer...

where a witness may frequently depose that in private, which he will be ashamed to testify in a public and solemn tribunal. There an artful or careless scribe may make a witness speak what he never meant, by 207. The Constitution itself, for example, refers to and relies on oaths in several key passages. See U.S. CONST. art. I, § 3, cl. 6 (requiring "Oath or Affirmation" when Senate sits as solemn court of impeachment); id. art. II, § 1, ci. 8 (elaborating presidential oath); id. art. VI, cl. 3 (requiring various officers and legislators to take an "Oath or Affirmation" to support the Constitution); id. amend IV (requiring "Oath or affirmation" for search or seizure warrants); see also Marbury v. Madison, 5 U.S. (1 Cranch.) 137, 180 (1803) (stressing the judicial oath); McCulloch, 17 U.S. (4 Wheat.) at 416 (authorizing Congress to add new oaths). For further discussion of the greater weight placed on testimony under oath than on unsworn statements, see Westen, supra note 82, at 86-87, 90-91, 100 & n.122, 111, 147. 208. In Jed Rubenfeld's terminology, our theory "captures" the "paradigm" case-in this case, trial by affidavit. See Jed Rubenfeld, Reading the Constitution as Spoken, 104 YALE L.J. 1119, 1169-71 (1995).

1996]

SIXTH AMENDMENT FIRST PRINCIPLES

dressing up his depositions in his own forms and language; but here he is at liberty to correct and explain his meaning, if misunderstood, which he can never do after a written deposition is once taken. Besides, [cross examination] will sift out the truth much better than a formal set of interrogatoriespreviouslypenned and settled.2 °9

In one of its earliest and most quoted expositions on the clause, the

Supreme Court echoed Blackstone: "The primary object of the [Confrontation Clause] was to prevent depositions or exparte affidavits... being used

against the prisoner in lieu of a personal examination and cross-examina210 tion of the witness. What's more, our reading of "witness" in the Confrontation Clause nicely meshes with the best reading of "witness" in its fraternal twin, the

Compulsory Process Clause. 211 A defendant should be able to oblige witnesses to take the stand at trial; but she should also be allowed to oblige pretrial depositions and affidavits, "canning" testimony to be later introduced in court in situations where the witness might not be available at the

time of the trial. (Imagine an alibi on his deathbed, pretrial.) When the shoe is on the other foot, the government has the power, pretrial, to subpoena a dying eyewitness 212 and "can" her affidavit; and as we shall see, the Compulsory Process Clause should give the accused subpoena parity. 213 (Note also how this "canning" may help the accused preserve 209. See 3 BLACKSTONE, supra note 52, at *373 (emphasis added); see also HALE, supra

note 142, at 345. 210. Mattox v. United States, 156 U.S. 237, 242 (1895). Recent quotations to this passage include: Maryland v. Craig, 497 U.S. 836, 845 (1990); Kentucky v. Stincer, 482 U.S. 730, 736-37 (1987); California v. Green, 399 U.S. 149, 157-58 (1970); Barber v. Page, 390 U.S. 719, 721 (1968); Douglas v. Alabama, 380 U.S. 415, 418-19 (1965). 211. Though I disagree with his approach in important respects, I share Professor Westen's views that (1) the key to the Confrontation Clause is the word "witness"; (2) the Confrontation and Compulsory Process Clauses are siblings; (3) an ideal interpretive theory should be able to read "witness" the same way in both clauses; and (4) once the word "witness" is properly read, it "can and should be taken literally." Peter Westen, The Future of Confrontation,77 MICH. L. REV. 1185, 1201-02 (1979). 212. As this word implies, a person who sees an underlying out-of-court event is in one ordinary-language sense a "witness"-but surely this alone cannot be the test for the Confrontation Clause. If she never declares anything, in court or out, she is not a Confrontation Clause witness even under the Court's test. More generally, even if the government gets a statement from this eyewitness pretrial, so long as her declarations are never alluded to at trial, surely she is not a Confrontation Clause "witness against [the accused]"; the government need not somehow bring her face to face with the defendant. See McCray v. Illinois, 386 U.S. 300, 313-14 (1967) (holding that Confrontation Clause does not require that government produce police informant to testify in court for cross-examination by defendant); Craig, 497 U.S. at 864-65 (Scalia, J., dissenting) (contending that "witness" refers to one who gives testimony at trial). (Of course the defendant may well want to subpoena her and use her testimony on the stand under the Compulsory Process Clause, and is free to do so.)

213. See infra Part IVc; cf. FED. R. CRIM. P. 15(a) (providing for defendant-initiated pretrial depositions in "exceptional circumstances"); Abraham S. Goldstein, The State and

696

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

evidence during a long accusation period, thus protecting one of the interests that we saw earlier in the speedy trial context.) In addition, the very existence of the Compulsory Process Clause also powerfully undercuts any possible fairness concern about our straightforward reading of "witness" in the Confrontation Clause. If witness A testifies about what out-of-court friend B said, and the defendant wants to challenge B's memory or truthfulness directly, face to face, the defendant can always use his own compulsory process right to subpoena B and interrogate him on the stand, for all to see. 21 4 Happily, our reading of the Confrontation Clause squares with the 15 results of almost all modern Supreme Court cases. 2 The Court has rightly seen that the word "witness" must go beyond those who take the stand in the flesh. But the Court, at least in its language, has failed to crisply distinguish between general out-of-court declarations--one friend talking to another, often even before the government is involved-and governmentally prepared depositions.2 16 In its results, however, the Court has intuitively sensed this difference. Of course, just because garden-variety hearsay that the government tries the Accused: Balance of Advantage in CriminalProcedure, 69 YALE L.J. 1149, 1180-92 (1960)

(noting various pretrial disparities between government and defendant, but not relying on Compulsory Process Clause); Westen, supra note 82, at 128-29 (analyzing same issue through lens of Compulsory Process Clause). 214. For similar reminders, see White v. Illinois, 502 U.S. 346, 355 (1992); United States v. Inadi, 475 U.S. 387, 397-99 & nn.7, 9 (1986). Obviously, the right to compulsory process must encompass, where appropriate, the right to treat the witness as "hostile," to crossexamine him with leading questions, and even to impeach his testimony: the very notion of compulsory process suggests the possibility of an obvious conflict of interest between the witness and the accused. Evidentiary rules that prohibit a defendant from impeaching his "own" witness, see, e.g., Chambers v. Mississippi, 410 U.S. 284, 295-98 (1973), violate the obvious spirit of the Compulsory Process Clause, and basic innocence-protecting and truthseeking principles to boot. For a thoughtful analysis, see Peter Westen, Confrontation and Compulsory Process:A Unified Theory of Evidence for Criminal Cases, 91 HARV. L. REV. 567, 601-13 (1978). 215. The one possible exception is the 5-4 ruling in Idaho v. Wright, 497 U.S. 805 (1990) (finding a Confrontation Clause violation when pediatrician testified about incriminating statements concerning sex abuse that were made to him by defendant's young daughter). Even here, however, the result might well be justifiable on my approach. For the reasons, see Brief for the United States as Amicus Curiae, White, 502 U.S. 346 (No. 90-6113) [hereinafter Brief for the United States]. 216. Two Justices, however, have properly drawn this distinction, accepting the views propounded by the United States, as amicus curiae. See White, 502 U.S. at 358-66 (Thomas, J., concurring, joined by Scalia, J.). But see id. at 352-53 (opinion of the Court, per Rehnquist, C.J.) (brushing aside this approach as coming "too late in the day" without considering how much more coherence it could offer to explain the results of past cases). Professors Friedman and Graham have also suggested that not all out-of-court declarants are Confrontation Clause "witnesses." Alas, both scholars sweep some friend B-like statements into their "witness" definitions. See Richard D. Friedman, Toward a PartialEconomic, Game-Theoretic Analysis of Hearsay, 76 MINN. L. REV. 723, 726 n.10 (1992); Michael H. Graham, The Confrontation Clause, The Hearsay Rule, and Child Sexual Abuse Prosecutions: The State of the Relationship, 72 MINN. L. REV. 523, 593-98 (1988).

1996]

SIXTH AMENDMENT FIRST PRINCIPLES

to introduce at trial clears the Confrontation Clause hurdle does not mean it is automatically admissible. The flexible hearsay rules, with their flexible exceptions, will typically stand as additional nonconstitutional barriersmuch like nonconstitutional statutes of limitations in the speedy trial context. And under an evolving hearsay doctrine, courts can focus directly on reliability, without having to bend the words or the rule-like grammar of the Confrontation Clause. Beyond this reliance on ordinary rules of evidence, if certain hearsay is horrendously unreliable-in a way the jury could not appropriately discount for-the truth-seeking and innocenceprotecting principles of the Due Process Clause and the Sixth Amendment more generally could come into play. And here we see the final reason for objecting to the Court's shotgun wedding of the hearsay rule and the Confrontation Clause.2 17 Sometimes hearsay will be reliable, and will help a defendant prove her innocence.2 18 It would be highly unfortunate if we read the Constitution as a document opposed to hearsay as such. For on such a reading, an innocent defendant might face tough sledding in trying to introduce certain reliable hearsay to prove his innocence, despite his right to put on the stand witnesses in his own favor.2 19 To the clause undergirding that right, the Confrontation Clause's fraternal twin, we now turn. C. COMPULSORY PROCESS

"[Tihe accused shall enjoy the right... to have compulsory process for obtaining witnesses in his favor." Underlying this clause is nothing less than the right of the accused to offer "his defence" at his "public trial"-to show the jury and the world that he is not guilty of the "infamous" and "criminal" conduct of which he

217. As the Solicitor General put the point in a brief outlining a theory of the Confronta-

tion Clause very similar to mine: "The right to confrontation is a feature of criminal

procedure intended to benefit the defendant. By contrast, the hearsay rule is a feature of the law of evidence applicable to all litigants in both civil and criminal proceedings." Brief for the United States, supra note 215, at 24; see also Dutton v. Evans, 400 U.S. 74, 97 n.4 (1970) (Harlan, J., concurring) (similar). 218. See, e.g., Chambers, 410 U.S. 284 (defendant sought admission of hearsay evidence of another man's confession to the murder charged against defendant). 219. Imagine, for example, a certain reliable subcategory of hearsay that the government seeks to introduce against a defendant. If the Court strikes down, on Confrontation Clause grounds, governmental efforts to introduce this type of hearsay, the government may respond by not letting defendants in other cases introduce this type of hearsay either. And this will

hurt some innocent defendants in other cases. These defendants may object to the hearsay rule, arguing that it violates their right to subpoena and put on their own witnesses under the Compulsory Process Clause. But the government can now respond by hiding behind parityneither side in any case can ever introduce this kind of hearsay. See infra Part IVc.

698

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

stands "accused." 22 This defense can begin with vigorous questioning of government witnesses and evidence under the letter and spirit of the Confrontation Clause; but it reaches full bloom in the Compulsory Process Clause right to present witnesses and evidence of his own. To be sure, the Clause explicitly speaks only of compelling witnesses, but surely the rights to present witnesses who volunteer, and to present physical evidence, follow a fortiori. If the accused, in order to show his innocence, is generally empowered to drag a human being, against her will, into the courtroom to tell the truth, surely he must also enjoy the lesser-included rights to present other truthful evidence that in no way infringes on another human being's autonomy. These lesser-included rights are plainly presupposed by the Compulsory Process Clause;.2 . the entire innocence-protecting and truth-seeking structure of the Sixth Amendment, and of constitutional criminal procedure generally, makes no sense without them. (And here we see again the difference between, on one hand, a plain meaning approach sensitive to letter and spirit and, on the other hand, wooden literalism.) At least three interesting sets of questions arise concerning the scope of the Compulsory Process Clause. First, does a defendant have a right to knowingly put on perjured testimony, if he thinks he can get away with it, fooling (at least for now) the jury and perhaps even the public? Second, exactly how much compulsion is a defendant entitled to use? Suppose a state enforces its subpoenas by fining those who defy. If a reluctant defense witness opts to pay the fine rather than testify, can a defendant insist that she be "compelled" even more-say, held in contempt? Or, if that is not enough, may a defendant insist that the government threaten to boil the witness in oil? Third, exactly whom may the defendant compel as a "witness?" Does the clause give a defendant a right to compel a recalcitrant husband to testify about marital conversations with his wife, or a reluctant doctor to testify, without patient consent, about a patient's private medical condition? What if the government has in place a general spousal or doctor-patient privilege scheme that would prevent the prosecution from compelling analogous witnessing? In light of all that we have seen, the first question seems easy. A major purpose of the Sixth Amendment, in its Public Trial and Confrontation Clauses, is to deter and detect perjury and other frauds upon the court: the Amendment seeks truth, not lies. A witness who breaks her oath may be punished for perjury; and a defendant or attorney who conspires with the witness may likewise be punished for soliciting or suborning this perjury. The Sixth Amendment gives the accused a right to show he did not engage in "infamous" conduct, not a right to perpetrate infamous conduct in the 220. See Westen, supra note 82, at 182-84. 221. Similarly, the Fourth Amendment's reference to Americans' right to be secure in their "persons, houses, papers, and effects" plainly presupposes background rights of tort and property law, and their accompanying remedies-trespass law and the like.

1996]

SIxTH AMENDMENT FIRST PRINCIPLES

699

2 22

courtroom itself. The answers to our second and third questions are related: A defendant should be given compulsion parity with the government. Whatever compulsion the government could use against a given recalcitrant witness, a defendant can use. If the government cannot compel a doctor to testify against his patient, because of a general doctor-patient privilege, a defendant cannot so compel this doctor. The constitutional virtues of this test are many. It offers a clean, easy-toadminister rule. It borrows from general themes elsewhere in constitutional law--of nondiscrimination and virtual representation-in areas like equal protection, and interstate privileges and immunities. 223 The idea is to enable defendants to benefit from the balance that the state tries to strike when its own evidence-seeking self-interest is at stake. In general, the government will want to give itself broad subpoena power. Without such power, government in a great many cases cannot hope to carry its heavy Winship burden of proof. And so when a government chooses to deny itself a certain coercion technique (threats to boil a reluctant witness in oil), or even all coercion against certain highly valued social relationships of intimacy and trust (like the one between wife and husband), this self-denial proves that the government really does see a "compelling interest" against compulsion. Though the words of the Compulsory Process Clause do not, on their face, demand a parity reading, the established Anglo-American right that the clause meant to declare was clearly defined in terms of subpoena parity. The landmark English Treason Act of 1696 gave defendants "the like Processe ...to compell their Witnesses ...as is usually granted to compell Witnesses to appeare against them."22' 4 Some early American formulations echoed this. Thus, William Penn's 1701 Pennsylvania Charter of Privileges declared that defendants "shall have the same Privileges of Witnesses ...as the Prosecutors; 2 2 5 and the New Jersey State Constitution of 1776 gave accused persons "the same privileges of witnesses ... as their prosecutors are or shall be entitled to."' 2 2 6 The precise federal formulation, using the language of "compulsory process," derives, it seems, from Blackstone's influential Commentaries, and here too, the parity idea is 222. See Nix v. Whiteside, 475 U.S. 157 (1986) (holding that Sixth Amendment right to assistance of counsel not violated when attorney refuses to cooperate with defendant in presenting perjured testimony). 223. See U.S. CONST. amend. XIV, § 1 (equal protection); id. art. IV, § 2, cl. 1 (interstate privileges and immunities); see also McCulloch v. Maryland, 17 U.S. (4 Wheat.) 316, 436 (1816) (state may tax real property of federal bank on same terms as it taxes like private real property); JOHN HART ELY, DEMOCRACY AND DISTRUST 77-87 (1980) (developing similar

themes). 224. 7 Will. 3, ch. 3, § 7 (1696) (emphasis added). 225. PA. CHARTER art. V (1701) (emphasis added). 226. N.J. CONST. art. XVI (1776) (emphasis added).

700

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

explicit: the accused must enjoy "the same compulsive process to bring in his witnesses for him, as was usual to compel their appearance against him."22 7 The first Congress, which drafted the Confrontation Clause in 1789, statutorily implemented its rule in 1790 in words that, once again, explicitly sounded in parity: "[A capital defendant] shall have the like process of the court ... to compel his ... witnesses to appear at his... trial, as is usually granted ... to appear on [behalf of] the prosecution."" In its sparse compulsory process caselaw, the Court has repeatedly 22 9 but in struck down asymmetric witness rules, and noted the asymmetry, one important corner of law, the government today can compel witnessing where defendants cannot. When government prosecutes defendant A, it will often need the testimony of A's (perhaps junior) partner-in-crime B in order to carry its Winship burden and make its charges stick. But when the government subpoenas B, B may claim his self-incrimination privilege. (Otherwise, the government could simply force B to testify in A's case, and then turn around and prosecute B, using a transcript of B's prior testimony against himself.) But the government may nonetheless compel B, even in the teeth of B's self-incrimination claim, so long as government gives B minimal Fifth Amendment immunity. According to the modern Court in the Kastigar2 30 case, this immunity must guarantee that the government will not use B's testimony, or any fruits derived from the testimony, in any later criminal case against B. The government may still prosecute B, but it must prove that everything in its case is wholly independent of B's earlier to prove, as the compelled testimony. This, of course, can be a hard thing 23 1 can attest. special prosecutor in the Oliver North case

227. 4 BLACKSTONE, supra note 52, at *351 (emphasis altered). On Blackstone as Madison's inspiration here, see Westen, supra note 82, at 97-98 & n.114. 228. Federal Crimes Act of 1790, ch. 9, 1 Stat. 112, 118 (emphasis added); see also 8 WIGMORE, supra note 145, § 2191 (stating that clause merely gives defendant "the common right.., possessed both by parties in civil cases.and by the prosecution in criminal cases"). 229. See Webb v. Texas, 409 U.S. 95, 96, 98 (1972) (noting that trial judge intimidated sole witness for defense, but not prosecution witnesses); Washington v. Texas, 388 U.S. 14, 22 (1967) (noting that accomplices were allowed to testify for government but not for defendants); see also id. at 24-25 (Harlan, J., concurring in the judgment) (stressing this fact); Pennsylvania v. Ritchie, 480 U.S. 39, 57 & n.14 (1987) (distinguishing between symmetric and asymmetric privileges in due process analysis); cf. Green v. Georgia, 442 U.S. 95, 97 (1979) (invalidating, on due process grounds, exclusion of hearsay statement that defendant sought to introduce, where government introduced same statement in another criminal proceeding); Cool v. United States, 409 U.S. 100, 103 n.4 (1974) (rejecting as "fundamentally unfair" instruction telling jury it could convict solely on basis of accomplice testimony but not telling jury it could acquit solely on this basis, in case where defendant put accomplice on the stand); Chambers v. Mississippi, 410 U.S. 284, 295-98 (1973) (invalidating, under Due Process Clause, trial in which defendant was barred from impeaching his own witness while government was free to impeach that witness). 230. Kastigar v. United States, 406 U.S. 441 (1972). 231. See United States v. North, 920 F.2d 940 (D.C. Cir. 1990), cert. denied, 500 U.S. 941 (1991) (amending United States v. North, 910 F.2d 843 (D.C. Cir. 1990)). For discussion, see Amar & Lettow, Fifth Amendment, supra note 3, at 875-80.

1996]

SIXTH AMENDMENT FIRST PRINCIPLES

But now suppose that the government decides not to compel B's testimony via minimal Fifth Amendment (Kastigar) immunity, but that defendant A wants to oblige B to take the stand. A believes that B's truthful testimony would in fact prove A's innocence. B did it, not A. A subpoenas B, but B resists, claiming his self-incrimination privilege. A now demands that the government give B Kastigar immunity so as to overcome B's privilege, and compel B to testify. What result? Under current doctrine, A would lose.2 32 Otherwise, every defendant could give his partners in crime an "immunity bath," and the government will be much worse off when it tries to prosecute the criminal partners in later cases. The government, again, would have to prove that nothing in its case was in any way influenced by the defendant's prior compelled testimony, and this can be very hard to prove-especially when witnesses testify as expansively as they can to maximize their future immunity. Note, however, what has happened. Defendant A is being denied his express right to compel B to witness-in testimony that might indeed prove A innocent. The minimal idea of subpoena parity underlying the Compulsory Process Clause is violated here: government can "compel" B to testify (via Kastigar) but the defendant may not. B's Fifth Amendment rights seem to be at war with, and to prevail against, A's Sixth Amendment rights, even though A is of course (for now) the one who is on trial, perhaps with his very life at stake. The problem here, however, has been created by the Court, not the Constitution. The Court has simply misread the word "witness"-again that key word!-in the Self-Incrimination Clause. And once we read that clause correctly, the Compulsory Process Clause springs free from its current shackles. As Ren6e Lettow and I have explained elsewhere,2 33 Kastigar is wrong. So long as the government never introduces B's transcript as testimony in a later criminal case against B, B will never have become an involuntary witness against himself in his own case. Fruits of prior testimony are allowable; fruits are not witnessing. And so the government may compel B's testimony in A's case merely by giving B "testimonial" immunity-B's testimony cannot itself be introduced against B. Under this rule, if the government decides not to immunize and compel B, A can do so. A can indeed enjoy compulsion parity, but the government need never fear immunity baths. When A immunizes and compels B, the government is never worse off than it would have been without immunity. Indeed, it is better off, for now it can use any leads or fruits generated by B to prosecute B. Through the Compulsory Process Clause, innocent A can in

232. See generally Peter W. Tague, The Fifth Amendment: If an Aid to the Guilty Defendant, an Impediment to the Innocent One, 78 GEO. L.J. 1 (1989) (collecting and analyzing lower

court cases). 233. See generallyAmar & Lettow, Fifth Amendment, supra note 3.

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

fact help the truth come to light and bring guilty B to justice. Instead of being at war with each other, the Fifth and the Sixth Amendments can now work together, in a way that promotes truth and protects innocence. Happier still, Fifth Amendment testimonial immunity would also powerfully reinforce Sixth Amendment Speedy Trial Clause values. Currently, many a defendant may languish pretrial while the government first tries to convict an accomplice, who can then be obliged to testify against the defendant without the high cost of Kastigar immunity. With testimonial immunity in place, the government could instead prosecute the defendant A and his alleged accomplice B together in a speedy joint trial, with two juries empanelled (one for each defendant). If either A or the government seeks to oblige B to take the stand, B's jury can be dismissed-this is, in effect, testimonial immunity-but A's jury can hear all. And the prosecutor or B, of course, could similarly compel A to testify, with A's jury excused, if A pleads the Fifth. On this reading, the Constitution coheres, with testimonial immunity rules tightly intermeshing with speedy trial values and compulsion parity ideas. Of course, even compulsion parity does not entirely even the scales between prosecution and defendant. Beyond its power to "compel" witnesses, government can "encourage" them to volunteer, by covering their expenses, or promising them greater immunity than is required to "compel" them under the Self-Incrimination Clause. Government can also spend vast sums of money to investigate crime. Most defendants lack a parity of resources to "encourage" and "investigate" other witnesses. The parity underlying the Compulsory Clause is in this sense a narrow one; it is 23 4 a parity of subpoena-like compulsion power. But the idea that parity is the touchstone only for compulsion can at times help defendants, especially innocent defendants. Consider a case where a defendant is not seeking to compel a balky witness, but trying to put on the stand a voluntary witness or to introduce some physical evidence. Assume that, under a general evidence rule in place in the jurisdiction-a rule applicable against prosecutors and defendants-this evidence is inadmissible. By hypothesis, the evidence rule passes a parity test, but the defendant can respond that parity only applies to compulsion. Here, the defendant can argue that because no unwilling human being is being coerced, the test should be a more general Sixth Amendment and due 234. Government may also execute surprise searches with search warrants. Though "compulsory" in a sense-a target must honor the warrant-these searches do not involve subpoena-like "process," and thus seem to fall outside the scope of the Compulsory Process Clause and the parity principle. Perhaps, however, if a defendant can show probable cause that a specific person has stolen goods or contraband in a specific place, and would destroy the stuff if served with a subpoena, then the government must execute a defendant-initiated search warrant under the parity principle. If prosecutors may use warrants when subpoenas would predictably be defied, see Amar, FourthAmendment, supra note 2, at 765-66, 779-80, perhaps defendants should have similar power to assure true subpoena parity.

1996]

SIXTH AMENDMENT FIRST PRINCIPLES

. 703

process test of innocence protection and truth-seeking: unless the evidence is so unreliable, in the context of other evidence in the case, that it cannot properly be assessed by the jury and the public, a defendant should be able to get it in.2 35 He has a general right to make his defense-to show he didn't do it-lest we unduly increase the risk of erroneously convicting an innocent man. It remains, finally, to ponder more carefully the plight of the innocent man who seeks to compel a witness who hides behind a general privilegesay, a doctor who refuses to testify about private medical facts concerning her patient, unless the patient consents. Unlike, say, a governmentinformer privilege, this one passes the parity test. Nevertheless, our defendant responds: "I am innocent. And this doctor has evidence-reliable evidence-that can prove my innocence to the jury and to the public." Though there appears to be little support for the defendant's plea in the history of the Compulsory Process Clause, its English roots, or its early implementation in America, it still tugs at our hearts. But perhaps its very emotional power is its own undoing; for at a minimum, the defendant (or his counsel) can make this speech to the jury, and argue that the unavailability of the evidence should raise reasonable doubts.23 6 Alternatively, we might say that compulsion parity is necessary to vindicate the Compulsory Process Clause, but not sufficient.2 3 7 If, however, we reject this approach, and embrace compulsion parity as both necessary and sufficient, we are admitting that the truth and innocence protection are not the only important values-we as a society also care about other things, including preserving fragile and socially beneficial relationships of trust like the bonds between wife and husband, doctor and patient, priest and penitent. But such a recognition does not demolish my general claims here about the huge importance of truth-seeking and innocence protection in the Sixth Amendment and in constitutional criminal procedure generally. After all, if innocence protection were the only value, to be maximized at all costs, we would simply refuse to allow anyone to be convicted, in order to eliminate any possibility of an erroneous conviction of an innocent man. 235. For a similar suggestion, see Westen, supra note 82, at 133-36, 156-57, 159. 236. See Alfred Hill, Testimonial Privilege and Fair Trial, 80 COLUM. L. REV. 1173, 1175 (1980). Consider also the case where a prosecution witness asserts a privilege in response to a defendant's vigorous cross-examination-a Confrontation Clause, rather than Compulsory Process Clause, context. If the privilege passes the parity test, a court could uphold the privilege but strike the witness's other testimony on the ground that the privilege may not be invoked selectively to present possibly misleading half-truths. See Davis v. Alaska, 415 U.S. 308 (1974) (invalidating conviction where government witness testified and then, on crossexamination, hid behind a half-truth about his juvenile arrest record-a half-truth defendant was not allowed to expose). Note, however, that Davis did not involve a true privacy privilege, see infra text accompanying notes 238-41, but instead involved a prosecution witness's attempt to exclude "public" information. 237. See Westen, supra note 82, at 173-77.

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

Moreover, the existence of true privacy privileges-doctor-patient, spousal, and priest-penitent, for example-surely does not support broad upsidedown exclusionary rules of constitutional criminal procedure, like the Fourth Amendment Boyd/Weeks/Mapp 238 rule and the Fifth Amendment Kastigar23 9 rule. To begin with, privacy privileges are simply not constitutional criminal procedure. Indeed, like Voltaire's Holy Roman Empire they are often none of the above. First, these privileges have typically not been entrenched as textual constitutionalmandates, and so if their precise contours prove inconvenient, they can be easily adjusted. Second, they are most definitely not criminal: they apply in all courtrooms, civil and criminal. By contrast, the Fourth Amendment exclusionary rule applies only in criminal and not civil cases; and the Kastigar rule likewise bars fruits only in criminal cases but not civil ones. Third, privacy privileges are not truly procedural. They are rooted in substantive norms, protecting valuable social relations-bonds of trust and privacy that would be destroyed by the public witnessing itself. Exclusionary rule rhetoric at times tries to sound the same themes, but this rings hollow. The underlying Fourth Amendment protects privacy, but the introduction in court of, say, drugs or stolen goods itself works no new privacy violation.2 4 ° (If it did, exclusion would be required in civil cases too.) Of course, friends of the Fourth Amendment exclusionary rule argue it will deter future violations, but that is very different from privacy privileges where exclusion is itself the right, not an analytically misfitting remedy. The analytic misfit of ordinary Fourth Amendment exclusion is that exclusion is neither necessary nor sufficient to deter. It is not sufficient because it fails to focus on cops who know a citizen is innocent, and predictably find no evidence in the course of hassling her. Once we realize, as did the Founders, that we must provide deterrent remedies for innocent citizens, we can put an analytically proper remedy in place, and exclusion is no longer necessary to deter. And, not coincidentally, our proper remedial scheme will be right side up, making innocent citizens whole and denying guilty defendants windfalls. In short, the Framers understood deterrence much better than do exclusionary rule fans.

238. See Mapp v. Ohio, 367 U.S. 643 (1961) (holding that evidence obtained by unconstitutional search is inadmissible in criminal case against searchee); Weeks v. United States, 232 U.S. 383 (1914) (similar); Boyd v. United States, 116 U.S. 616 (1886) (similar). 239. See Kastigar v. United States, 406 U.S. 441 (1972). 240. In rare situations, the evidentiary introduction of certain illegally obtained items might itself be a privacy violation-say, reading a woman's diary in open court. But typically this privacy violation does not depend on the illegality of a prior governmental search. Reading the diary in open court, civil or criminal, would be a privacy violation even if the government lawfully obtained the diary-in a proper search, or by subpoena, or if handed to the government by some third-party thief. Although the exclusionary rule is inapplicable in all these situations, a true privacy-based analysis might allow for a "diary privilege" in all proceedings, criminal or civil. See Amar and Lettow, Fifth Amendment, supra note 3, at 921.

1996]

SIXTH AMENDMENT FIRST PRINCIPLES

What's more, true privacy privileges respect the deep norms underlying a public trial; they simply mark the border between the public and the truly private. As the word privilege implies, these immunities shield private communications that have never before been made public. And so when these privileges are observed, we do not suffer the public demoralization that now occurs, when publicly available information is simply shut out of the public courtroom.2 4 ' The public trial ideal promises that anyone who seeks to offer reliable evidence can "draw nigh, and will be heard." The exclusionary rule betrays this ideal; but many privacy privileges do notthey do not bar members of the public from coming to court and volunteering their information. D. COUNSEL

"[T]he accused shall enjoy the right ... to have the Assistance of Counsel for his defence." This last right is a big one, ramifying in many directions. Here I shall briefly discuss only three. First, when does the right attach? Second, does it encompass the right to state-provided counsel? Third, what are the ethical limits of counsel's role in assisting the accused in making her defense? In answering these questions, we shall see how the clause neatly interlocks with other clauses in the Sixth Amendment; here too, innocence protection and truth-seeking are, or should be, central. Given the explicit words of the Counsel Clause, and the overall architecture of the Sixth Amendment, the "right ... to have the Assistance of Counsel for [one's] defence" is triggered when and because a person is "accused" of "criminal" wrongdoing. As we saw in our discussion of the Speedy Trial Clause, the Sixth Amendment is accusation-based,because accusation itself subjects a person to distinct risks.24 2 One risk is the threat of prolonged pretrial detention-a threat triggered by arrest or indictment. To prevent this evil, an accused may need to file a habeas writ; to compensate for any past violation and deter any future one, an accused may need to bring a Bivens-like action. And these are things that lawyers are (or at least should be) good for. Filing writs and the like requires 241. So too, under First Amendment doctrine, it is one thing to prevent the press from ever gaining access to a certain bit of information, and a very different-and far more problematic-thing to prevent the press from publishing that same bit of information if the press already knows of it. See Gannett Co. v. DePasquale, 443 U.S. 368, 393 n.25 (1979) (distinguishing between prohibition of access to information about a criminal trial and prohibition of such information already in one's possession); id. at 411, 447 (Blackmun, J., concurring in part and dissenting in part) (distinguishing between press's right of access to judicial proceedings and press's right to print information learned in open court). 242. See Kirby v. Illinois, 406 U.S. 682, 689-90 (1972) (plurality opinion) (suggesting that all Sixth Amendment guarantees are accusation-based); see also supra Part II.

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

familiarity both with broad substantive law-the rights of free men and women-and technical procedural law: where to file, with what words, before whom, and so on. And so for pretrial detainees, the right to counsel simply makes real one of the core legal interests guaranteed by the Speedy Trial Clause. Even for an "accused" not subject to arrest or pretrial detention, the formal accusation itself of course triggers another threat to liberty. Formally, indictment sets the stage for trial; once indicted, even a person who honestly pleads not guilty risks erroneous conviction. In order to defend himself, an accused needs to understand the accusation-this is the obvious logic of the "nature and cause" clause. But indictments can at times be laced with technical legal language that an accused person-especially, perhaps, if wholly innocent (say, in a case of mistaken identity)-may not understand. Here too, the "Assistance of Counsel" can help breath life into the promise of an earlier Sixth Amendment clause. Consider, finally, the trial itself, when the accused at last gets a chance to clear his name before the jury and the world-to poke holes in the government's case and to present his own, revealing the truth and showing his factual or normative innocence. But this trial may be filled with technical lawyer's law-for example, the rules of evidence. And so without expert legal assistance, the accused may well be unable to exercise his all-important rights to cross-examine government witnesses, and to present his own evidence and witnesses-rights at the heart of the Confrontation and Compulsory Process Clauses. Once again, we see how the Sixth Amendment's last clause merely makes real the promise of its earlier provisions.24 3 Of course, there may well be occasions prior to criminal accusation when general, innocence-protecting principles will trigger an analogous right of counsel, but there is no blanket rule requiring counsel before accusation. 2 " Consider, for example, the immemorial practice of excluding defense lawyers from the grand jury inquest. 245 An arrest may trigger the risk of extended pretrial detention, but a grand jury summons does not. A trial is governed by technical rules of evidence, but a grand jury is not.

243. For similar analysis tightly linking the counsel right to legal complexity, see, e.g., United States v. Gouveia, 467 U.S. 180, 188-89 (1984); Gagnon v. Scarpelli, 411 U.S. 778, 789 (1973); Kirby, 406 U.S. at 689-90; Coleman v. Alabama, 399 U.S. 1, 9-10 (1970); Johnson v. Zerbst, 304 U.S. 458, 462-63 (1938); Powell v. Alabama, 287 U.S. 45, 69 (1932). Note also that the Treason Act of 1696, the font of the right of counsel, spoke of "[c]ounsel learned in the law." 7 & 8 Will. 3, ch. 3, § 1. 244. See Moran v. Burbine, 475 U.S. 412, 428-30 (1986) (finding no Sixth Amendment right to counsel during interrogation in police station prior to accusation, and rejecting contrary language of Miranda v. Arizona, 384 U.S. 436 (1966) and Escobedo v. Illinois, 378 U.S. 478 (1964)). 245. For a functional defense of this established practice, see Amar & Lettow, Fifth Amendment, supra note 3, at 899 n.192.

1996]

SIxTH AMENDMENT FIRST PRINCIPLES

Thus, the Sixth Amendment analysis we have developed explains both why a person has a general right of counsel after and because of accusation, and why she has no blanket right before, even if she is paying the bill. But what about those "accused" persons who cannot pay the bill for the "Assistance of Counsel?" Are they entitled to counsel at government expense? The text of the Counsel Clause can be read either way. But structural arguments strongly support the modern Court's view that indigent defendants have a right to the assistance of counsel at government expense.2 46 As we have seen, in many respects the last clause of the Sixth Amendment simply implements the rights guaranteed by its earlier clauses, and these clauses protect rich and poor alike. The appointment of counsel requires government to act "affirmatively," but so does the Compulsory Process Clause, which requires government to act affirmatively to enforce subpoenas. Government payment of defense counsel costs money, but so does convening an impartial jury, holding a public trial, informing the accused of the "nature and cause" of charges, and so on. At first, history might seem to strongly undercut our textual and structural analysis. The very Congress that proposed the Sixth Amendment provided, in its statutes, for appointed counsel in capital cases but not elsewhere.2 4 7 But this history should not be overread. The legal theory was not that indigents in noncapital cases would simply have to do without the "Assistance of Counsel." Rather, it was that the judge himself could act as counsel and provide a defendant legal assistance.2"' Because judges were of course paid by the government, this too was a form of governmentprovided "counsel" for indigent defendants. But, of course, this was a plainly inadequate form, especially because some of the rights in the Sixth Amendment are based on a healthy suspicion of judges. And the need for expert legal advice has only grown over the last two centuries, as criminal law-both substantive and procedural-has become increasingly complex. Even if the history of the first Congress were (wrongly) deemed to trump text and structure in the Sixth Amendment context, the indigent's right to appointed counsel could also be derived from the innocence-protecting spirit of the Due Process Clause. The flexibility of the clause, focusing explicitly on how much process is due, can easily accommodate evolving historical developments. Twentieth-century America is considerably wealthier than was eighteenth-century America;, and this in turn bears on 246. See, e.g., Johnson, 304 U.S. at 462-63 (noting that Sixth Amendment right to be heard would often be of little avail if it did not include right to assistance of counsel); Gideon v. Wainwright, 372 U.S. 335, 344-45 (1963) (same). 247. See Federal Crimes Act of 1790, ch. 9, 1 Stat. 112, 118. 248. See, e.g., John H. Langbein, The Historical Origins of the Privilege Against SelfIncrimination at Common Law, 92 MICH. L. REV. 1047, 1050-52 (1994) (describing this

theory).

THE GEORGETOWN LAW JOURNAL

.[Vol. 84:641

whether additional procedural safeguards, though costly, are nonetheless due-reasonable, apt, fair.2 49 In today's world, an indigent defendant without counsel runs an undue risk of being convicted, even if wholly innocent. The importance of innocence and truth also helps answer our final question, concerning the appropriate ethical boundaries of counsel's role. As we have seen, a guilty defendant has no general right, under the Sixth Amendment or anything else, to perjure himself, or knowingly introduce false evidence, or commit other infamous acts in the courtroom itself. The point here is not merely that if he does commit a fraud upon the Court, and is found out, he may be punished. Rather, it is that he may not do it, even if he knows he can "get away with it," or is willing to "run the risk." This "get away with it" mentality-that a person has a "right" to commit perjury (or murder witnesses for that matter) so long as he can cover his tracks, or is willing to "do the time"-reflects a Holmesian "bad man" view of the law. But our Sixth Amendment is structured around a good man, wrongly charged. The "bad man" view is built on lies and concealment; but the Sixth Amendment was written to promote truth and openness-publicity. All of this has huge implications for the proper role of counsel. If the accused has no right to defraud the court, surely he has no right to the "assistance" of counsel in perpetrating his frauds. Bribing judges, lying on the stand, putting on false witness, fabricating evidence, threatening prosecution witnesses-none of these is part of a lawful Sixth Amendment "defence" and so counsel must avoid aiding any of these acts. Counsel, under the Sixth Amendment, is provided to enforce the law's letter and spirit, not to evade them. Thus, the Sixth Amendment's structure rejects the ideas that a lawyer owes some unbending duty of loyalty to the accused, and that a lawyer must never act against his client.2 5 ° Rather, counsel's obligations of client loyalty are limited by the truth-seeking architecture of the Sixth Amendment. To be sure, effective "Assistance of Counsel" cannot occur if a 249. At one time, the precise doctrinal basis for the right to appointed counsel-the Counsel Clause or the Due Process Clause-mattered a great deal; today it matters not. In the late 1930s the Court located the right not in due process, but in the Sixth Amendment. See Johnson, 304 U.S. at 462. As a result, states were not bound by the Johnson rule since the

Sixth Amendment was then viewed as applying only to federal trials. A Fifth Amendment due process rationale in Johnson, by contrast, would have applied to states by dint of the Fourteenth Amendment's parallel Due Process Clause. In the 1960s, however, the provisions of the Sixth Amendment were held applicable against states; and so today, little turns on the Due Process Clause/Counsel Clause distinction. See generally Henry J. Friendly, The Bill of Rights as a Code of Criminal Procedure, 53 CAL. L. REV. 929, 942-46 (1965) (explaining interplay of incorporation debate and appointed counsel doctrine). 250. For an example of "unbending loyalty" ideology, see Monroe H. Freedman, Professional Responsibility of the CriminalDefense Lawyer: The Three Hardest Questions, 64 MIcH. L. REV. 1469 (1966).

1996]

SixTH AMENDMENT FIRST PRINCIPLES

lawyer is merely the spy of the system. And so, the attorney-client relationship is shielded by a privacy privilege akin to the husband-wife, doctorpatient, and priest-penitent privileges. But the attorney-client privilege is also different. Spouses, doctors, and priests may simply stay out of the courtroom, and keep their private information out of the public trial. The lawyer, by contrast, plays a hugely public and active role; simple nonfeasance is not feasible, and so he must take care to avoid becoming an active agent of fraud. In its landmark ruling in the 1986 case, Nix v. Whitside,25 1 the Supreme Court held that "under no circumstances may a lawyer either advocate or passively tolerate a client's giving false testimony. This, of course, is consistent with the governance of trial conduct in what we have long called 'a search for truth.' ,252 The logic of Nix is that perjury must be prevented, and not merely by the adversary system of prosecutorial cross-examination and the subsequent threat of perjury prosecution. Defense counsel herself must not assist this fraud, and merely putting a witness on the stand and asking questions counts as impermissible assistance. In short, at times a lawyer cannot ask questions and let the jury decide, even if this strategy maximizes her client's chance for victory. "Plainly, [counsel's duty of loyalty] is limited to legitimate, lawful conduct compatible with the very '25 3 nature of a trial as a search for truth. This logic, however, raises unsettling questions about extremely vigorous cross-examination of truthful witnesses, a tactic that most American defense lawyers today view as obviously appropriate and perhaps required by legal ethics. 2 54 Consider a bank robbery case in which the accused tells his lawyer that he, in fact, committed the crime and that the security guard saw him. If the guard takes the stand as a prosecution witness, how vigorously may defense counsel cross-examine her? Does the Sixth Amendment oblige, permit, or prohibit questions on cross-examination to show that the guard's eyesight is not great, that she only had a few seconds to observe the robber, and that she failed to immediately pick out the defendant in a line-up? A critic of this cross-examination might view it as the moral equivalent of fraud: Isn't counsel implicitly suggesting, through his questions, that the security guard is mistaken-a fact that counsel knows is false? Counsel, however, has a strong response: "My questions do not necessarily imply that my client is in fact innocent, but only that the government has failed to provide proof beyond a reasonable doubt. The

251. 475 U.S. 157 (1986). 252. Id. at 171. 253. Id. at 166. 254. Nix involved Compulsory Process Clause principles, which are implicated when a defendant puts on his own witnesses; but as we have seen, these principles tightly intertwine with Confrontation Clause principles implicated by cross-examination of government witnesses.

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

guard's eyesight and shaky identification do raise reasonable doubt, and so the implicit theory of the case underlying my admittedly vigorous questions is utterly truthful. Even if my client is in fact guilty, the government's proof is still weak, and this weakness should be emphasized to the jury and the public, in keeping with the true spirit of Sixth Amendment innocence protection." Counsel's line here is a fine one-between asking questions that imply factual innocence and asking questions that merely imply reasonable doubt-but, in principle, a workable one. But now consider a date rape case where the defendant, once again, tells his lawyer that he did it. May the lawyer try to demolish the victim when she takes the witness stand, impugning her motives, questioning her honesty, bringing up her past sexual history? In at least four ways, this case seems different, and the vigorous cross-examination at issue may actually tend to undermine the spirit of the Sixth Amendment. First, counsel, by his questions, is implicitly saying that the witness is not merely possibly mistaken, but is in fact a liar-indeed, a perjurer.2 55 In ordinary morality, there is a big difference between asking someone "are you sure?" and calling her a liar. In any other context, this knowingly false public accusation of dishonesty would be actionable, but when this reputational mugging occurs in a courtroom it is immune from defamation laws. Counsel can thus "get away with it" in a Holmesian "bad man" sense; but as we have seen the Sixth Amendment is built on a very different worldview. (Indeed, as we saw in the Speedy Trial context, the Amendment was designed to shield Americans from baseless reputational assaults, especially when defamation actions are unavailing because of courtroom immunities.) Second, savaging a truthful witness's honor and character powerfully discourages truthful witnesses from coming forward-in precise violation of the spirit of the Public Trial Clause. Outright extortion, bribery, or physical intimidation of a would-be witness is clearly illegitimate; but is a brutal verbal assault carried out in a public courtroom and built on lies really so different? (And let us not forget that this cruel verbal assault takes aim at a woman who has already suffered a savage physical assault from the lawyer's client.) Third, in trying to paint the rape victim as a liar, a lawyer typically must "play act" and posture in ways that implicitly argue to the jury and the public not merely that reasonable doubt about guilt exists, but that the accused is in fact not guilty. And the implicit representation is a knowing falsehood-a lie. 256 Fourth, and related, the emphasis on reasonable doubt in the bank robbery case closely tracks the core purpose of counsel in the Sixth Amendment, based on lawyers' special familiarity 255. Cf. Stephen A. Saltzburg, Lawyers, Clients, and the Adversary System, 37 MERCER L.

REV. 647, 676 (1986) ("The lawyer should not use her courtroom experience and the nervousness of the witness.., to make an honest witness appear less than honest."). 256. See David Luban, Are Criminal Defenders Different?, 91 MICH. L. REV. 1729, 1759-62 (1993) (discussing play acting).

1996]

SIXTH AMENDMENT FIRST PRINCIPLES

with legal rules, like the Winship principle. Anyone can insult, bully, play act, posture, and mislead-that is not why the Sixth Amendment calls for lawyers. Rather, lawyers contribute something special to a trial precisely because they understand the importance of procedural rules like Winship. It is no answer to all this to simply point to, and hide behind, the "adversary system" as a justification for trying to demolish a rape victim in the witness stand. As we have seen, the entire premise of the Sixth Amendment, and of the Confrontation Clause in particular, was that the truth will likely out in a vigorous exchange. But the question now at hand is precisely what the ground rules of that exchange should be so as to vindicate this premise, and maximize the possibility that the truth will indeed out. If perjury is beyond the pale-even when one could get away with it within an adversary system-perhaps certain forms of cross-examination (and other lawyers' tactics) should themselves be cross-examined. And because any ethical restrictions on misleading cross-examination would constrain prosecutors as well as defense counsel, such25 7restrictions in the long run may prove a net plus to innocent defendants. These final speculations will no doubt strike many American lawyers, steeped in gladitorial ethics and a sporting theory of justice, as outright heresy. But perhaps this reaction is itself only a sign of how far we have strayed from the Sixth Amendment's first principles. This Amendment speaks to and about lawyers with more directness than any other clause in the Constitution, and yet many lawyers today misread its letter and spirit. And so our current lack of a good map of the Amendment may mean not merely bad constitutional law and bad criminal procedure, but perhaps bad legal ethics too. CONCLUSION

This article completes a trilogy of essays designed to restore coherence to modern constitutional criminal procedure. Current caselaw under the Fourth, Fifth, and Sixth Amendments is often perverse. The words and spirit of the Constitution are too often ignored, the truth too often sup257. These ethical restrictions need not take the form of rules enforced by external sanction; they could instead simply stand as norms of appropriate professional conduct internalized by members of the bar. A lawyer bent on evading the spirit of ethical standards can often do so by pretending to hear no evil from his client and, where inconvenient facts come to light, by asking cynically (along with Pontius Pilate) "what is truth?" But I write here not about a bad-man lawyer, but about a good-woman lawyer, who seeks to do the right thing. English practice is far closer to my ideal, and so were ABA standards only twenty years ago. Compare AMERICAN BAR ASSOCIATION PROJECT ON STANDARDS FOR CRIMINAL JUSTICE, STANDARDS RELATING TO THE ADMINISTRATION OF CRIMINAL JUSTICE 132 (1974) (The Defense Function § 7.6(b)) (defense lawyer "should not misuse the power of crossexamination or impeachment by employing it to discredit or undermine a witness if he knows the witness is testifying truthfully") with AMERICAN BAR ASSOCIATION STANDARDS FOR CRIMINAL JUSTICE § 4-7.6 (2d ed. 1986 Supp.) (abandoning earlier standard).

712

THE GEORGETOWN LAW JOURNAL

[Vol. 84:641

pressed, the guilty too often rewarded, and the innocent too often victimized. As we confront the collective corpus of constitutional criminal procedure caselaw handed down by the Warren, Burger, and Rehnquist Courts, we see at many levels complexity, confusion, and sometimes contradiction. To fix the mess, we need to locate proper first principles of constitutional criminal procedure. I submit that a sound constitutional jurisprudence should begin with respect for the words of the Constitution, and a sound criminal procedure should begin with a commitment to protect the innocent and pursue the truth. In this trilogy, I have tried to show the consequences of this approach and, I hope, its many virtues.